Last 7 Days (August 29 – September 04, 2026)
Competitive programming has become a key test of large language model reasoning, with international competitions such as IOI and ICPC representing its most challenging settings. We present an end-to-end specialization pipeline combining large-scale problem curation, synthetic reasoning traces, supervised fine-tuning (SFT), and reinforcement learning (RL). Using 22,000 curated problems, we train Nemotron-3-Nano-CC (30B-A3B) with SFT and RL and Nemotron-3-Ultra-CC (550B-A55B) with SFT alone. We further introduce GenCorrect, a feedback-driven test-time compute strategy that iteratively generates, evaluates, and refines diverse solutions. On IOI 2025, Nano-CC improves from 130 points to 291 after post-training and to 468 with GenCorrect, exceeding the gold threshold of 438.3 while Ultra-CC reaches 502. Guided by these results, we develop a competition-specific Ultra-CC system and evaluate it prospectively during IOI 2026. Under the same time, internet-access, and submission constraints as human contestants, it scores 535.4 out of 600, exceeding both the gold threshold of 361.12 and the top human score of 498.27. To our knowledge, this is the first AI system to outscore the highest-scoring human contestant on an IOI problem set.
Primary: NVIDIA
All Institutions: NVIDIA
The paper presents a post-training pipeline and test-time strategy that enables an AI system to outperform top human contestants in the International Olympiad in Informatics. This represents a significant milestone in AI reasoning, demonstrating that through specialized curation, reinforcement learning, and iterative refinement, LLMs can achieve gold-medal performance in competitive programming, thereby establishing a new benchmark for complex algorithmic reasoning.
The paper proposes a comprehensive end-to-end pipeline for specializing large language models (LLMs) for competitive programming. The methodology combines large-scale problem curation (22,000 problems), synthetic reasoning trace generation, Supervised Fine-Tuning (SFT), and Reinforcement Learning (RL). A key methodological contribution is "GenCorrect," a feedback-driven test-time compute strategy that iteratively generates, evaluates, and refines diverse solutions. The approach is applied to two model scales: Nemotron-3-Nano-CC (30B-A3B) using SFT+RL, and Nemotron-3-Ultra-CC (550B-A55B) using SFT alone. The pipeline demonstrates that even smaller models can achieve competitive performance through effective post-training and test-time scaling.
The experimental evaluation is rigorous and high-stakes, utilizing the International Olympiad in Informatics (IOI) 2025 and 2026 as benchmarks. On IOI 2025, the Nano-CC model improved from a baseline of 130 points to 291 after post-training, and further to 468 with GenCorrect, exceeding the gold threshold. The Ultra-CC model reached 502 points. Most significantly, the paper reports a prospective evaluation on IOI 2026, where the Ultra-CC system scored 535.4 out of 600, surpassing both the gold threshold (361.12) and the top human score (498.27). This is a landmark result, claiming to be the first AI system to outscore the highest-scoring human contestant on an IOI problem set.
The paper provides detailed descriptions of the curation process, the SFT and RL training procedures, and the GenCorrect algorithm. However, as is common with frontier lab papers, specific hyperparameters, exact dataset compositions, and code may not be fully released. The use of proprietary NVIDIA Nemotron models limits immediate reproducibility for external researchers without access to these specific base models.
The primary limitation is the reliance on proprietary base models (Nemotron-3 series), which hinders independent verification and adoption by the broader community. Additionally, the evaluation is heavily focused on competitive programming, which may not generalize directly to other domains of reasoning or coding tasks. The "prospective" nature of the IOI 2026 evaluation, while impressive, raises questions about potential overfitting to the specific problem distribution if the training data overlapped with the test set, although the paper claims distinct constraints.
This paper has significant broader impact by demonstrating that LLMs can not only match but exceed human performance in highly complex, long-horizon reasoning tasks like competitive programming. The introduction of GenCorrect as a test-time compute strategy offers a generalizable technique for improving LLM performance without increasing model size. This result shifts the benchmark for AI reasoning capabilities and provides a roadmap for developing more specialized, high-performance AI systems for complex problem-solving. The paper presents a post-training pipeline and test-time strategy that enables an AI system to outperform top human contestants in the International Olympiad in Informatics. This represents a significant milestone in AI reasoning, demonstrating that through specialized curation, reinforcement learning, and iterative refinement, LLMs can achieve gold-medal performance in competitive programming, thereby establishing a new benchmark for complex algorithmic reasoning.
We introduce LLaDA-Image, a unified framework that pairs a 6B Diffusion Transformer (DiT) trained from scratch with a frozen vision-language understanding module built on the LLaDA2.0-Mini diffusion language model backbone. Instead of relying heavily on paired image-text data from the beginning, we first build a strong visual generative prior through image-only pre-training and mid-training. The generation pipeline comprises 220M samples, 98 of which are real images. For efficient and scalable optimization, we use parameter-free RMSNorm throughout the DiT together with the Muon optimizer. The resulting unified model produces highly photorealistic images while accurately following fine-grained editing instructions. We further distill LLaDA-Image into LLaDA-Image-Turbo, enabling fast inference in 2-4 sampling steps. On Qwen-Image-Bench, LLaDA-Image achieves overall scores of 53.53 and 53.38 on the English and Chinese tracks, respectively, setting a new state-of-the-art among open-source models on both tracks. To support further research on capable and efficient generative models, we release our model weights, training code, and detailed recipes.
Primary: Meituan
All Institutions: Meituan
The paper introduces LLaDA-Image, a unified open-source image generation framework that achieves state-of-the-art performance by decoupling visual prior learning from text-conditioning and employing efficient optimization techniques. This work is significant for its comprehensive open-sourcing of training recipes and its novel integration of diffusion-based language models for vision-language understanding, providing a robust baseline for future research in scalable and efficient image generation.
The paper proposes LLaDA-Image, a unified framework combining a 6B parameter Diffusion Transformer (DiT) with a frozen vision-language understanding module based on the LLaDA2.0-Mini diffusion language model. The core methodological contribution lies in the training recipe: rather than relying immediately on paired image-text data, the authors first establish a strong visual generative prior through image-only pre-training and mid-training. This decoupling of visual prior learning from text-conditioning is a significant architectural and training strategy choice. The use of parameter-free RMSNorm and the Muon optimizer for scalable optimization of the DiT is a practical engineering contribution that addresses efficiency concerns in large-scale diffusion model training. The integration of a diffusion-based language model (LLaDA) as the understanding backbone, rather than a standard autoregressive LLM, is a novel architectural choice that aligns the generative and understanding components within a similar probabilistic framework, potentially reducing modality gaps.
The evaluation is conducted on Qwen-Image-Bench, a benchmark specifically designed for evaluating image generation quality and instruction following. The model achieves state-of-the-art scores among open-source models (53.53 English, 53.38 Chinese). The inclusion of a distilled version, LLaDA-Image-Turbo, which enables 2-4 step inference, is a strong practical contribution, demonstrating that high quality can be maintained with significant speedups. The claim of "highly photorealistic images" and "accurately following fine-grained editing instructions" is supported by the benchmark scores, though a direct comparison with closed-source models (like DALL-E 3 or Midjourney) is absent, which is typical for open-source releases but limits the full scope of the SOTA claim. The data pipeline involving 220M samples with 98% real images suggests a robust data curation strategy, though specific details on the filtering and balancing of the remaining 2% are not fully detailed in the abstract.
The paper explicitly states the release of model weights, training code, and detailed recipes. This is a high standard for reproducibility. The use of standard components (DiT, RMSNorm, Muon) combined with open-sourcing of the specific training pipeline makes it highly reproducible for other labs. The "fully open training recipes" claim is a major positive factor, as many recent large-scale models release weights but not the training data or hyperparameters.
The primary limitation is the reliance on a specific benchmark (Qwen-Image-Bench) which may not capture all aspects of image generation quality, such as aesthetic appeal or safety. The 6B parameter size is substantial, and while the Turbo version helps, the base model's computational cost for inference and training remains high. The "frozen" nature of the vision-language module means that the model's understanding capabilities are capped by the LLaDA2.0-Mini backbone, and any improvements in the understanding module would require retraining or fine-tuning the entire system. Additionally, the lack of comparison with other recent open-source DiT-based models (like SD3 or PixArt) in the abstract makes it difficult to gauge the relative improvement in the open-source landscape specifically.
The release of a strong, open-source image generator with detailed training recipes has significant positive impact on the research community. It enables further research on efficient training, data curation, and distillation techniques for diffusion models. The use of a diffusion language model for understanding is an interesting direction that could inspire further work on unifying generation and understanding in non-autoregressive frameworks. The potential for misuse in generating deepfakes or disinformation is a standard concern for image generators, but the open nature of the model allows for the development of detection and mitigation tools by the community. The paper introduces LLaDA-Image, a unified open-source image generation framework that achieves state-of-the-art performance by decoupling visual prior learning from text-conditioning and employing efficient optimization techniques. This work is significant for its comprehensive open-sourcing of training recipes and its novel integration of diffusion-based language models for vision-language understanding, providing a robust baseline for future research in scalable and efficient image generation.
We present Alignment-Free Text-Audiobox (Text-AB), a unified framework for high-quality voice dubbing and full-duplex dialogue synthesis. Building on a Diffusion Transformer trained with a flow-matching objective, Text-AB departs from the Audiobox system along three dimensions. First, it operates in a latent diffusion framework using DAC-VAE features that encode 48 kHz waveforms into a 25 Hz latent sequence, giving over 10x higher compression than previous EnCodec representations while improving resynthesis quality. Second, Text-AB is alignment-free: it consumes raw text via an off-the-shelf text encoder and learns text-speech alignment through cross-attention, removing the need for forced alignment and explicit duration prediction. Third, we scale model and data substantially, pretraining a 3B-parameter model on 480k hours of monolingual speech, followed by supervised fine-tuning on three downstream tasks: cross-lingual voice dubbing, full-duplex dialogue synthesis, and emotional full-duplex dialogue synthesis. At inference, Text-AB supports one-shot generation for up to ~1 min of speech and arbitrarily long-form generation via a multi-diffusion scheme, plus a multi-stage reranking strategy that enhances quality based on automated metrics. On a real-world dubbing benchmark, Text-AB delivers a step-change improvement over the latest internal dubbing system, with large gains in prosody similarity, voice similarity, naturalness, and shareability. For full-duplex dialogue synthesis, it approaches human recordings on short-form conversations and substantially outperforms the latest internal model on long-form human-likeness and expressivity, while natively modeling turn-taking, back-channeling, and emotional dynamics. For emotional dialogue synthesis, emotion conditioning significantly improves emotion alignment and emotional interaction quality over the unconditioned baseline.
Primary: Meta AI
All Institutions: Meta AI
The paper presents a highly effective, scalable, and unified framework for advanced speech synthesis tasks, demonstrating state-of-the-art performance in voice dubbing and full-duplex dialogue generation through the combination of latent diffusion, alignment-free text conditioning, and large-scale training.
The paper proposes Text-AB, a unified framework for voice dubbing and full-duplex dialogue synthesis. The core methodological contributions are: (1) shifting from EnCodec to DAC-VAE latent features for higher compression and fidelity; (2) an "alignment-free" architecture that uses raw text embeddings via cross-attention instead of forced alignment and explicit duration prediction; and (3) a multi-stage training pipeline (pretraining on monologues, SFT for dubbing, SFT for dialogue) on a 3B parameter Diffusion Transformer. The use of flow-matching is standard in recent high-fidelity TTS (e.g., VoiceBox, Audiobox), and the latent diffusion approach is also established. The novelty lies primarily in the integration of these components into a single scalable system that handles both single-speaker dubbing and two-channel full-duplex dialogue without architectural changes other than input/output projection adjustments. The "alignment-free" aspect is a significant practical improvement, removing a brittle dependency on forced aligners.
The evaluation is extensive, covering three distinct tasks: cross-lingual voice dubbing, full-duplex dialogue synthesis, and emotional dialogue synthesis. The results show substantial improvements over internal baselines, particularly in human-likeness and prosody similarity. The use of both objective metrics (WER, SpkSim, Aesthetics) and subjective human evaluations (MOS, pairwise comparisons) provides a robust assessment. The ablation studies on model size, training data, and inference strategies (multi-diffusion, reranking) are thorough and demonstrate clear scaling laws and the importance of the proposed inference techniques. The comparison to ground-truth data for short-form dialogue is particularly compelling, showing near-human parity.
The paper provides detailed descriptions of the architecture, training data composition (hours, languages), hyperparameters (learning rates, batch sizes, GPU counts), and inference settings (ODE steps, reranking size). However, as is common with large-scale industrial papers, the specific datasets are internal and not publicly released. The code is not mentioned as being released. The reliance on internal benchmarks limits external reproducibility, but the methodological details are sufficient for a skilled practitioner to attempt a reproduction with similar data.
The primary limitation is the lack of public code and data, which hinders independent verification. The model is large (3B parameters) and requires significant computational resources for training and inference (256 A100s for training, multiple candidates for reranking). The evaluation is largely comparative against internal baselines, making it difficult to compare directly with other state-of-the-art open-source systems. The "alignment-free" approach relies on a powerful text encoder (mT5), which adds complexity and potential latency.
This work has significant implications for the development of full-duplex speech language models (FD-SLMs), which are a key component of next-generation voice assistants. By providing a high-quality, controllable method for generating synthetic full-duplex dialogue data, Text-AB can help address the data scarcity problem in training FD-SLMs. The unified framework for dubbing and dialogue synthesis also has practical applications in media production and accessibility. The alignment-free design simplifies the TTS pipeline, making it more robust to noisy or conversational speech. The paper presents a highly effective, scalable, and unified framework for advanced speech synthesis tasks, demonstrating state-of-the-art performance in voice dubbing and full-duplex dialogue generation through the combination of latent diffusion, alignment-free text conditioning, and large-scale training.
We introduce LLaDA-Image, a unified framework that pairs a 6B Diffusion Transformer (DiT) trained from scratch with a frozen vision-language understanding module built on the LLaDA2.0-Mini diffusion language model backbone. Instead of relying heavily on paired image-text data from the beginning, we first build a strong visual generative prior through image-only pre-training and mid-training. The generation pipeline comprises 220M samples, 98 of which are real images. For efficient and scalable optimization, we use parameter-free RMSNorm throughout the DiT together with the Muon optimizer. The resulting unified model produces highly photorealistic images while accurately following fine-grained editing instructions. We further distill LLaDA-Image into LLaDA-Image-Turbo, enabling fast inference in 2-4 sampling steps. On Qwen-Image-Bench, LLaDA-Image achieves overall scores of 53.53 and 53.38 on the English and Chinese tracks, respectively, setting a new state-of-the-art among open-source models on both tracks. To support further research on capable and efficient generative models, we release our model weights, training code, and detailed recipes.
Primary: Meituan
All Institutions: Meituan
The paper introduces LLaDA-Image, a unified open-source image generation framework that achieves state-of-the-art performance by decoupling visual prior learning from text-conditioning and employing efficient optimization techniques. This work is significant for its comprehensive open-sourcing of training recipes and its novel integration of diffusion-based language models for vision-language understanding, providing a robust baseline for future research in scalable and efficient image generation.
The paper proposes LLaDA-Image, a unified framework combining a 6B parameter Diffusion Transformer (DiT) with a frozen vision-language understanding module based on the LLaDA2.0-Mini diffusion language model. The core methodological contribution lies in the training recipe: rather than relying immediately on paired image-text data, the authors first establish a strong visual generative prior through image-only pre-training and mid-training. This decoupling of visual prior learning from text-conditioning is a significant architectural and training strategy choice. The use of parameter-free RMSNorm and the Muon optimizer for scalable optimization of the DiT is a practical engineering contribution that addresses efficiency concerns in large-scale diffusion model training. The integration of a diffusion-based language model (LLaDA) as the understanding backbone, rather than a standard autoregressive LLM, is a novel architectural choice that aligns the generative and understanding components within a similar probabilistic framework, potentially reducing modality gaps.
The evaluation is conducted on Qwen-Image-Bench, a benchmark specifically designed for evaluating image generation quality and instruction following. The model achieves state-of-the-art scores among open-source models (53.53 English, 53.38 Chinese). The inclusion of a distilled version, LLaDA-Image-Turbo, which enables 2-4 step inference, is a strong practical contribution, demonstrating that high quality can be maintained with significant speedups. The claim of "highly photorealistic images" and "accurately following fine-grained editing instructions" is supported by the benchmark scores, though a direct comparison with closed-source models (like DALL-E 3 or Midjourney) is absent, which is typical for open-source releases but limits the full scope of the SOTA claim. The data pipeline involving 220M samples with 98% real images suggests a robust data curation strategy, though specific details on the filtering and balancing of the remaining 2% are not fully detailed in the abstract.
The paper explicitly states the release of model weights, training code, and detailed recipes. This is a high standard for reproducibility. The use of standard components (DiT, RMSNorm, Muon) combined with open-sourcing of the specific training pipeline makes it highly reproducible for other labs. The "fully open training recipes" claim is a major positive factor, as many recent large-scale models release weights but not the training data or hyperparameters.
The primary limitation is the reliance on a specific benchmark (Qwen-Image-Bench) which may not capture all aspects of image generation quality, such as aesthetic appeal or safety. The 6B parameter size is substantial, and while the Turbo version helps, the base model's computational cost for inference and training remains high. The "frozen" nature of the vision-language module means that the model's understanding capabilities are capped by the LLaDA2.0-Mini backbone, and any improvements in the understanding module would require retraining or fine-tuning the entire system. Additionally, the lack of comparison with other recent open-source DiT-based models (like SD3 or PixArt) in the abstract makes it difficult to gauge the relative improvement in the open-source landscape specifically.
The release of a strong, open-source image generator with detailed training recipes has significant positive impact on the research community. It enables further research on efficient training, data curation, and distillation techniques for diffusion models. The use of a diffusion language model for understanding is an interesting direction that could inspire further work on unifying generation and understanding in non-autoregressive frameworks. The potential for misuse in generating deepfakes or disinformation is a standard concern for image generators, but the open nature of the model allows for the development of detection and mitigation tools by the community. The paper introduces LLaDA-Image, a unified open-source image generation framework that achieves state-of-the-art performance by decoupling visual prior learning from text-conditioning and employing efficient optimization techniques. This work is significant for its comprehensive open-sourcing of training recipes and its novel integration of diffusion-based language models for vision-language understanding, providing a robust baseline for future research in scalable and efficient image generation.
The problem. A long-lived KV cache must be compressed before the queries that will read it exist; selection by observed attention (H2O, SnapKV) collapses there (0.00-0.33 needle retrieval on a NoPE MLA model), because a token's importance has not yet been observed. The method. On Kimi Linear, VestigeKV evicts by a query-independent signal the cache already carries: the 64-dimensional decoupled branch, a vestige of RoPE that NoPE training repurposes into a salience channel. Reading 11% of each row, it partitions the cache: the top-m rows stay in the attended tier; every other row moves -- exactly, never deleted -- to a GPU-resident archive reachable per step by a certified trigger. No training, no quantization, no weight or kernel change. Cost. Nothing measurable: retrieval holds at 1.00 under 8x and 0.92 under 32x from 8k to 65k context, zero gap to full-row selection. The attended tier is 0.25 KB of Kimi Linear's 8.1 KB per-token cache at 32x; the archive stays bit-exact and GPU-resident, with host offload as the VRAM-reclaiming variant. The recall tier -- the standard configuration -- holds 128x at 1.00. Kimi K3 is reported to use a NoPE Gated-MLA variant; if its cache layout matches, the method plausibly extends there -- we make no claim beyond the measured model. NoPE exclusivity. The identical operator on a RoPE MLA collapses to 0.08 (plain eviction: 0.42); query-independent salience itself exists only without rotation (top-1 targets span 2.3-6.7% of tokens vs. 10.2-46.8%), and query-universal exact merging is provably impossible under RoPE. All thresholds were frozen before data; 20 archived verdicts and 8 closed routes accompany the paper.
Primary: Anonymous (Double-Blind Review)
All Institutions: Anonymous
VestigeKV leverages the query-independent salience signal inherent in the decoupled branch of NoPE-MLA caches to enable lossless KV cache compression up to 32x (and 128x with recall) without training or quantization. The paper provides a rigorous theoretical foundation for why this signal exists only in NoPE architectures and demonstrates significant empirical improvements over standard attention-based eviction methods, offering a novel path for efficient long-context inference in specific model families.
The paper proposes VestigeKV, a KV cache eviction strategy for NoPE-MLA models (specifically Kimi Linear). The core insight is that in NoPE (No Positional Encoding) architectures, the "decoupled branch" of the MLA cache, which previously carried positional information, becomes a query-independent salience channel. The authors prove that under NoPE, attention scores are determined by a fixed bilinear form, allowing for query-independent ranking of token importance. This enables a "pre-query" eviction strategy where tokens are ranked by their intrinsic salience (read from the 64-dim branch) rather than observed attention. The method partitions the cache into an "attended tier" (top-m rows) and a "GPU-resident archive" (rest of rows), with a recall tier that fetches archived rows if a certified trigger indicates they are relevant. The theoretical contribution includes proving that exact merging of cache rows is impossible under RoPE but possible under NoPE for identical content, and that query-independent salience exists only in the absence of rotation.
The experiments are conducted on Kimi Linear 48B. The primary metric is needle retrieval intact rate. The paper reports that VestigeKV maintains 1.00 retrieval accuracy at 8x compression and 0.92 at 32x compression, outperforming H2O and SnapKV which collapse to 0.00-0.33. The "recall tier" extends this to 128x compression with 1.00 retrieval. The paper also reports perplexity and MAUVE scores, showing minimal degradation. The evaluation includes a comparison with a "RoPE MLA" baseline where the same operator fails (0.08 retrieval), highlighting the NoPE exclusivity. The experimental setup includes a serving-form engine reproduction to ensure practical viability.
The paper provides a strong reproducibility statement, mentioning pre-registration files, run records (JSON), and scripts to regenerate figures. It states that raw activation dumps are excluded for size but can be regenerated from public checkpoints. The use of pre-registered decision rules and known-bad input tests for gates adds to the rigor.
The method is specific to NoPE-MLA architectures (Kimi Linear). It is unclear if it extends to other NoPE variants or standard RoPE models (the paper explicitly states it collapses on RoPE). The evaluation is limited to one model family (Kimi Linear 48B) and one task (needle retrieval/perplexity), though the authors acknowledge this. The "mechanism of the depth" is noted as open. The paper relies on the specific cache layout of Kimi Linear, and extension to other models like Kimi K3 is plausible but untested.
This paper offers a significant insight into the internal structure of NoPE-MLA models, revealing that the "vestigial" branch carries salience information. This could influence future architectural designs for efficient inference. The method provides a practical, training-free way to compress KV caches significantly without loss, which is highly relevant for long-context LLMs. However, its applicability is limited to models that adopt the NoPE-MLA design, which is currently a niche but growing area. VestigeKV leverages the query-independent salience signal inherent in the decoupled branch of NoPE-MLA caches to enable lossless KV cache compression up to 32x (and 128x with recall) without training or quantization. The paper provides a rigorous theoretical foundation for why this signal exists only in NoPE architectures and demonstrates significant empirical improvements over standard attention-based eviction methods, offering a novel path for efficient long-context inference in specific model families.
On-policy distillation (OPD) combines student-generated rollouts with dense token-level supervision from a teacher. Existing work has mainly studied its algorithmic behavior, leaving the role of training data unclear. We examine this role at the data-minimal limit by training on a single query. One-shot OPD keeps improving for hundreds of steps and recovers most of full-data OPD's gain across task domains and model families. We explain this result through the states visited during training and the rate at which the student aligns with the teacher. We measure \emph{state coverage}, the fraction of the states full-data OPD visits that a query set's rollouts reach. A single query already reaches \(71.5\%\), most of it within the first 100 steps. Adding semantically distinct queries raises coverage and validation accuracy together, until 16 queries reach \(98.9\%\) and match full-data training. Yet alignment slows at a similar pace whether OPD trains on one query or the whole dataset, and even a fixed set of states takes hundreds of steps to absorb. OPD is therefore data-overfed but algorithm-starved. Its rollouts quickly expose broad supervision, while the student absorbs that supervision increasingly slowly. The state-coverage result extends to multi-teacher OPD, where 16 semantically diverse queries per domain match full-data MOPD. As a further stress test, content-light templates and off-domain WildChat queries also approach the real-query baseline. Task content and induced state coverage can therefore come apart. We hope these findings direct future work toward the step efficiency of OPD, and prompt a re-examination of the data and the mechanisms behind its recent successes in frontier post-training.
Primary: Northeastern University
All Institutions: Northeastern University
The paper demonstrates that On-Policy Distillation is highly data-efficient, with a single query covering 71.5% of the state space and 16 queries matching full-data performance, revealing that the bottleneck is algorithmic absorption speed rather than data volume. This finding challenges the standard assumption that distillation requires large datasets and provides a new metric, state coverage, to understand the dynamics of on-policy learning.
The paper investigates the data efficiency of On-Policy Distillation (OPD) by pushing the training data limit to a single query. The core methodological contribution is the introduction of "state coverage" as a metric to explain why minimal data suffices. The authors demonstrate that a single query generates rollouts that visit 71.5% of the states visited by full-data OPD, and that 16 semantically distinct queries reach 98.9% coverage. The analysis further decouples data coverage from algorithmic absorption, showing that while states are covered quickly, the student's alignment with the teacher slows down over time regardless of data volume. This "data-overfed but algorithm-starved" characterization provides a mechanistic explanation for the observed data efficiency.
The experiments are rigorous in their controlled setting, comparing one-shot, few-shot (up to 16 queries), and full-data OPD across multiple task domains and model families. The results are consistent: one-shot OPD recovers most of the full-data gain, and adding queries improves accuracy in tandem with state coverage. The inclusion of multi-teacher OPD (MOPD) and stress tests with content-light templates and off-domain WildChat queries strengthens the claim that state coverage, rather than specific task content, drives the learning. The finding that 16 queries match full-data performance is a strong empirical result.
The paper provides sufficient detail on the definition of state coverage and the experimental setup (model families, task domains) to allow for replication. However, without explicit code release or detailed hyperparameter schedules for the "hundreds of steps" mentioned, exact reproduction of the training dynamics may be challenging. The metric itself is clearly defined, making it reproducible in principle.
The primary limitation is the scope of the "single query" experiment; it is unclear if this extreme data minimality holds for more complex, long-horizon tasks where state space might be less saturated by a single rollout. Additionally, the paper focuses on the mechanism of data efficiency but does not propose a new algorithm to address the "algorithm-starved" aspect (i.e., speeding up absorption), leaving the practical implication as a call for future work rather than a solution.
This paper has significant implications for the cost of LLM post-training. If 16 diverse queries can approximate full-data distillation, the computational cost of generating rollouts for distillation could be reduced by orders of magnitude. It shifts the focus of the community from data scaling to algorithmic efficiency in on-policy learning, potentially influencing how frontier labs design their post-training pipelines. The paper demonstrates that On-Policy Distillation is highly data-efficient, with a single query covering 71.5% of the state space and 16 queries matching full-data performance, revealing that the bottleneck is algorithmic absorption speed rather than data volume. This finding challenges the standard assumption that distillation requires large datasets and provides a new metric, state coverage, to understand the dynamics of on-policy learning.
Competitive programming has become a key test of large language model reasoning, with international competitions such as IOI and ICPC representing its most challenging settings. We present an end-to-end specialization pipeline combining large-scale problem curation, synthetic reasoning traces, supervised fine-tuning (SFT), and reinforcement learning (RL). Using 22,000 curated problems, we train Nemotron-3-Nano-CC (30B-A3B) with SFT and RL and Nemotron-3-Ultra-CC (550B-A55B) with SFT alone. We further introduce GenCorrect, a feedback-driven test-time compute strategy that iteratively generates, evaluates, and refines diverse solutions. On IOI 2025, Nano-CC improves from 130 points to 291 after post-training and to 468 with GenCorrect, exceeding the gold threshold of 438.3 while Ultra-CC reaches 502. Guided by these results, we develop a competition-specific Ultra-CC system and evaluate it prospectively during IOI 2026. Under the same time, internet-access, and submission constraints as human contestants, it scores 535.4 out of 600, exceeding both the gold threshold of 361.12 and the top human score of 498.27. To our knowledge, this is the first AI system to outscore the highest-scoring human contestant on an IOI problem set.
Primary: NVIDIA
All Institutions: NVIDIA
The paper presents a post-training pipeline and test-time strategy that enables an AI system to outperform top human contestants in the International Olympiad in Informatics. This represents a significant milestone in AI reasoning, demonstrating that through specialized curation, reinforcement learning, and iterative refinement, LLMs can achieve gold-medal performance in competitive programming, thereby establishing a new benchmark for complex algorithmic reasoning.
The paper proposes a comprehensive end-to-end pipeline for specializing large language models (LLMs) for competitive programming. The methodology combines large-scale problem curation (22,000 problems), synthetic reasoning trace generation, Supervised Fine-Tuning (SFT), and Reinforcement Learning (RL). A key methodological contribution is "GenCorrect," a feedback-driven test-time compute strategy that iteratively generates, evaluates, and refines diverse solutions. The approach is applied to two model scales: Nemotron-3-Nano-CC (30B-A3B) using SFT+RL, and Nemotron-3-Ultra-CC (550B-A55B) using SFT alone. The pipeline demonstrates that even smaller models can achieve competitive performance through effective post-training and test-time scaling.
The experimental evaluation is rigorous and high-stakes, utilizing the International Olympiad in Informatics (IOI) 2025 and 2026 as benchmarks. On IOI 2025, the Nano-CC model improved from a baseline of 130 points to 291 after post-training, and further to 468 with GenCorrect, exceeding the gold threshold. The Ultra-CC model reached 502 points. Most significantly, the paper reports a prospective evaluation on IOI 2026, where the Ultra-CC system scored 535.4 out of 600, surpassing both the gold threshold (361.12) and the top human score (498.27). This is a landmark result, claiming to be the first AI system to outscore the highest-scoring human contestant on an IOI problem set.
The paper provides detailed descriptions of the curation process, the SFT and RL training procedures, and the GenCorrect algorithm. However, as is common with frontier lab papers, specific hyperparameters, exact dataset compositions, and code may not be fully released. The use of proprietary NVIDIA Nemotron models limits immediate reproducibility for external researchers without access to these specific base models.
The primary limitation is the reliance on proprietary base models (Nemotron-3 series), which hinders independent verification and adoption by the broader community. Additionally, the evaluation is heavily focused on competitive programming, which may not generalize directly to other domains of reasoning or coding tasks. The "prospective" nature of the IOI 2026 evaluation, while impressive, raises questions about potential overfitting to the specific problem distribution if the training data overlapped with the test set, although the paper claims distinct constraints.
This paper has significant broader impact by demonstrating that LLMs can not only match but exceed human performance in highly complex, long-horizon reasoning tasks like competitive programming. The introduction of GenCorrect as a test-time compute strategy offers a generalizable technique for improving LLM performance without increasing model size. This result shifts the benchmark for AI reasoning capabilities and provides a roadmap for developing more specialized, high-performance AI systems for complex problem-solving. The paper presents a post-training pipeline and test-time strategy that enables an AI system to outperform top human contestants in the International Olympiad in Informatics. This represents a significant milestone in AI reasoning, demonstrating that through specialized curation, reinforcement learning, and iterative refinement, LLMs can achieve gold-medal performance in competitive programming, thereby establishing a new benchmark for complex algorithmic reasoning.
Autonomous agents are beginning to carry out machine-learning (ML) research end to end. These agents combine a model backbone with a harness for planning, execution, memory, and verification, but this architecture still leaves domain-specific know-how outside the agent. We call this missing layer operational knowledge, the know-how that separates knowing a method from making it work. That knowledge is not absent from the field. It appears in repositories and papers, but in forms written for human readers and too large to load during a task. Once distilled into compact, verified skills, this knowledge can be reused across tasks rather than rediscovered during each run. We present DisCo, a skill-powered research agent that creates skills and uses them during research. Its distillation runs in two complementary forms: task-agnostic, condensing the field's widely used repositories into reusable skills, and task-oriented, producing the skills a concrete task calls for. The former, applied across the open ecosystem, yields the AREX-Skill Library, with 5,000+ verified skills distilled from 1,000 widely used ML repositories and organized into 20 areas and 178 capability families. With the GPT-5.5 backbone, research harness, and downstream execution budget held fixed, the skill-equipped research agent scores 134.3% higher on MLE-bench, 34.4% higher on PaperBench, 9.2% higher on FrontierCS, and 14.0% higher on PassNet than the same agent without skills. These gains come from adding distilled operating context under that fixed setup.
Primary: University of Science and Technology of China
All Institutions: University of Science and Technology of China, Beijing Academy of Artificial Intelligence, Renmin University of China
The paper introduces a skill distillation framework that significantly enhances autonomous ML research agents by providing reusable operational knowledge. The reported performance gains, particularly the 134.3% improvement on MLE-bench, demonstrate that bridging the gap between theoretical method knowledge and practical execution know-how is a critical lever for improving agent autonomy, making this a highly significant contribution to the emerging field of AI-for-AI.
The paper proposes "DisCo," a framework for distilling operational knowledge from GitHub repositories into compact, reusable "skills" for autonomous ML research agents. The methodology involves two distillation modes: task-agnostic (condensing widely used repositories into a static library) and task-oriented (generating skills specific to a current task). The core contribution is the AREX-Skill Library, containing over 5,000 verified skills from 1,000 repositories. The approach addresses the "operational knowledge gap" where agents know methods theoretically but lack the practical know-how to execute them without trial-and-error.
The authors report significant performance gains on four benchmarks: MLE-bench (134.3% higher), PaperBench (34.4% higher), FrontierCS (9.2% higher), and PassNet (14.0% higher) compared to a baseline agent without skills. The evaluation controls for the model backbone (GPT-5.5), research harness, and execution budget, isolating the impact of the skill layer. The magnitude of improvement on MLE-bench is particularly striking and suggests that operational context is a major bottleneck in current autonomous research agents.
The paper claims to provide a verified skill library and specific agent configurations. However, the full text provided is a skeleton with section headers but lacks the detailed implementation specifics, prompt engineering strategies, and verification protocols required to fully reproduce the skill distillation process. The reliance on a specific, potentially proprietary or rapidly evolving backbone (GPT-5.5) may also limit immediate reproducibility for the broader community.
The primary limitation is the dependency on the quality and coverage of the underlying GitHub repositories; if the source code is poor or non-standard, the distilled skills may be flawed. Additionally, the "verification" of skills is not detailed in the abstract, raising questions about how errors in the source code are filtered out. The approach may also struggle with highly novel tasks that have no existing repository precedent.
This work has high potential impact on the field of autonomous AI research. By formalizing "operational knowledge" as a distinct, reusable asset, it could enable agents to scale more effectively across diverse ML tasks. The creation of a large-scale skill library (AREX-Skill) could become a standard resource for agent developers, similar to how Hugging Face models are used for LLMs. The paper introduces a skill distillation framework that significantly enhances autonomous ML research agents by providing reusable operational knowledge. The reported performance gains, particularly the 134.3% improvement on MLE-bench, demonstrate that bridging the gap between theoretical method knowledge and practical execution know-how is a critical lever for improving agent autonomy, making this a highly significant contribution to the emerging field of AI-for-AI.
LLMs are increasingly deployed as orchestrators that coordinate specialized subagents to solve complex tasks through natural language. However, in many important domains like game playing and robotics, the strongest available agents are not language models. Integrating non-language agents with LLMs would require verbalization: compressing their rich continuous representations into sparse textual summaries at each interaction step. To study whether verbalization constitutes a bottleneck, we introduce LLAMIA-Bench, a suite of six diverse collaborative chess tasks spanning three facets: behavioral imitation, state assessment, and natural-language explanation. Each task instantiates a well-established chess problem that neither the LLM nor the chess engine can solve alone. To solve LLM collaboration with non-language agents, we introduce latent state internalization, which projects the subagent's continuous representations directly into the LLM's token stream as learned state tokens, with dynamic re-encoding as actions advance the environment state. Comparing internalization to verbalized integration, our experiments reveal a consistent verbalization debt: the performance gap widens throughout training and persists as the LLM scales from 4B to 14B parameters. A single 14B model, LLAMIA, trained with latent state internalization, matches or exceeds task specialists and frontier models including GPT-5.1 with tool access across all benchmark tasks, and generalizes out-of-distribution where task-specific finetunes collapse
Primary: Unknown
All Institutions: Unknown
The paper introduces latent state internalization to overcome the verbalization bottleneck in LLM-non-LLM agent collaboration, demonstrating superior performance over text-based integration in a chess benchmark. The work provides a compelling argument against the necessity of natural language as the sole interface for multi-agent systems, offering a technical pathway to preserve rich continuous information from specialized agents within LLM orchestrators.
The paper proposes "latent state internalization," a method that bypasses the natural language bottleneck in LLM-agent collaboration. Instead of forcing a non-language agent (like a chess engine) to verbalize its state into text, the method projects the agent's continuous internal representations directly into the LLM's token stream as learned "state tokens." This involves a dynamic re-encoding mechanism that updates these tokens as the environment state changes. The approach is technically sound and addresses a critical architectural limitation in current multi-agent systems where information loss occurs during verbalization. The design of the state tokens and their integration into the transformer architecture is a significant methodological contribution.
The authors introduce LLAMIA-Bench, a suite of six collaborative chess tasks covering behavioral imitation, state assessment, and explanation. The experimental setup compares the proposed internalization method against verbalized integration baselines. Key findings include a "verbalization debt" where the performance gap widens with training and model scale (4B to 14B). The 14B LLAMIA model reportedly matches or exceeds GPT-5.1 with tool access. While the results are impressive, the evaluation is confined to chess, which, while complex, is a single domain. The generalization to other domains (robotics, etc.) is claimed but not empirically demonstrated in the provided text.
The paper provides a project URL with code, data, and examples, which is a strong indicator of reproducibility. The specific details of the state token projection and re-encoding are described in the methodology, allowing for potential replication. However, without access to the specific hyperparameters and training protocols in the full text (which is truncated), full reproducibility is assumed based on the provided links.
The primary limitation is the domain specificity of the evaluation. Chess is a well-understood, deterministic game with a strong existing baseline (engines). It is unclear if the "verbalization debt" and the efficacy of latent internalization hold in stochastic, high-dimensional, or partially observable environments like robotics or real-world navigation. Additionally, the comparison to GPT-5.1 is a strong claim that requires careful scrutiny of the tool access setup to ensure a fair comparison.
If the findings generalize, this work could significantly impact the design of hybrid AI systems that combine LLMs with specialized non-LLM agents. It challenges the prevailing paradigm of natural language as the universal interface for agent communication, suggesting that direct latent space integration is superior for performance-critical tasks. This could lead to more efficient and capable multi-agent systems in robotics, autonomous driving, and complex simulation. The paper introduces latent state internalization to overcome the verbalization bottleneck in LLM-non-LLM agent collaboration, demonstrating superior performance over text-based integration in a chess benchmark. The work provides a compelling argument against the necessity of natural language as the sole interface for multi-agent systems, offering a technical pathway to preserve rich continuous information from specialized agents within LLM orchestrators.
People increasingly use language models to support life decisions. Many such decisions involve a probabilistic forecast: How likely is a major life event, a natural disaster, or an economic outcome? Users of language models may implicitly trust that these forecasts fall out of a coherent world model. In this paper, we evaluate the coherence of language model probabilistic forecasts through a procedure that builds on a theorem due to de Finetti. We elicit forecasts from language models across events generated from stock returns data. We then use linear programs to compute the largest Dutch-book profit - the profit an arbitrageur could guarantee by betting against model-generated probabilities - which we use as a measure of incoherence. Our procedure does not require outcome labels, so we can evaluate coherence even in settings where outcomes are not observed or have not yet resolved. We find substantial evidence of incoherence in language model forecasts. Such incoherence increases when there are richer logical relationships between events, and irrelevant contextual details can increase incoherence by an order of magnitude. We conclude by discussing how alternative training strategies may improve probabilistic coherence.
Primary: University of Chicago
All Institutions: University of Chicago, MIT, NBER
The paper introduces a rigorous, label-free evaluation framework for probabilistic coherence in LLMs using de Finetti's theorem and linear programming to compute Dutch book profits. By applying this method to stock return forecasting across 15 models and 365,100 elicitations, it reveals substantial and variable incoherence, demonstrating that logical consistency is a distinct and critical property of model forecasts that is sensitive to prompt design and logical complexity, thereby providing a new diagnostic tool for improving the reliability of probabilistic AI systems.
The paper proposes a rigorous, label-free evaluation framework for probabilistic coherence in Large Language Models (LLMs) based on de Finetti's theorem. By formulating the problem as a linear programming task to compute the maximum "Dutch book" profit (arbitrage gain), the authors create a metric that quantifies the distance of a model's stated probabilities from the set of coherent probability distributions. This approach is mathematically sound, leveraging the duality between the primal arbitrage problem and the dual distance-to-coherence problem. The novelty lies not in the mathematical theorem itself, which is classical, but in its systematic application to LLM evaluation using complex, logically structured event panels derived from financial data. The method effectively isolates logical consistency from predictive accuracy, offering a unique diagnostic tool for model reliability.
The experimental design is robust and extensive, involving 15 different LLMs and 365,100 total elicitations across 100 stock-days. The use of stock return data provides a structured domain with known logical relationships (complements, intersections, unions), which is crucial for testing coherence. The paper demonstrates that incoherence is prevalent and varies significantly across models, with some models showing up to 100x more incoherence than others. Key findings include the sensitivity of coherence to prompt details (e.g., irrelevant context increasing incoherence by an order of magnitude) and the improvement in coherence when models are forced to reason over the full event panel simultaneously. The correlation between coherence and accuracy is also explored, showing that while they are related, coherence is a distinct and more variable property.
The paper provides high reproducibility standards. It specifies the exact data sources (CRSP, Refinitiv), the sampling strategy, the specific models and providers used, and the prompt templates. The use of an open-weight model (GPT-OSS-120B) as the workhorse for detailed ablation studies enhances replicability. The code for the linear programming solver and data processing is implied to be available or straightforward to implement given the detailed description of the LP formulation. The inclusion of confidence intervals and bootstrap methods for statistical inference further strengthens the rigor of the results.
A primary limitation is the domain specificity; the findings are based on stock return forecasting, and it is unclear how well these results generalize to other domains like natural language understanding or scientific prediction. The paper acknowledges that the coherence ordering across models might not be invariant to different elicitation conditions. Additionally, the reliance on API-based models for some experiments introduces potential variability due to provider-specific settings (e.g., temperature, reasoning defaults) that are not fully controllable. The paper also notes that the "profits" are theoretical measures of incoherence rather than actual financial returns, which might limit direct practical interpretation for traders.
This paper has significant implications for the deployment of LLMs in high-stakes decision-making contexts where probabilistic forecasts are used. By providing a label-free metric for coherence, it enables developers and users to assess the logical consistency of model outputs without requiring ground truth data, which is often unavailable in real-time forecasting scenarios. The findings highlight the need for new training strategies, such as process reward models or RL objectives over groups, to improve the coherence of LLM forecasts. This work bridges the gap between classical decision theory and modern AI, offering a principled approach to evaluating and improving the reliability of probabilistic AI systems. The paper introduces a rigorous, label-free evaluation framework for probabilistic coherence in LLMs using de Finetti's theorem and linear programming to compute Dutch book profits. By applying this method to stock return forecasting across 15 models and 365,100 elicitations, it reveals substantial and variable incoherence, demonstrating that logical consistency is a distinct and critical property of model forecasts that is sensitive to prompt design and logical complexity, thereby providing a new diagnostic tool for improving the reliability of probabilistic AI systems.
Reinforcement learning is a natural way to post-train LLM agents for long-horizon interactive tasks judged only by end-of-task verification, yet a shared belief holds that outcome-only RL soon hits a ceiling on small open models. Recent work therefore compensates around the training with denser rewards, SFT priors, skill libraries, curated memory, or multi-agent orchestration. We argue the ceiling is an artifact of two failures of common practice. Signal starvation: group-relative RL with sparse outcome-only rewards yields a gradient only when a task's rollout group mixes successes and failures, so under-scaled exploration silences exactly the hardest, most instructive tasks. Policy drift: squeezing many updates out of a small task pool degrades the policy itself, as an unanchored objective lets the sampling distribution collapse exactly when saturation has already made informative groups rare. We present CANOPY (Coverage-ANchored On-PolicY RL), a minimalist protocol attacking both directly: scale same-task exploration until the natural signal reappears, keep every update on-policy, KL-anchored, and confined to the agent's own action tokens, then cash in an enlarged interaction budget at test time. On AppWorld, a long-horizon interactive coding benchmark, a Qwen3-14B policy trained with CANOPY through environment interaction alone--without task-specific supervision, auxiliary credit signals, or elaborate agent scaffolding--topped the public leaderboard (Feb. 2026; Test-Normal TGC 86.9, Test-Challenge 67.6), and the same design principles lift Qwen3.5-9B on SWE-bench Verified by 16.6 points. Agentic RL alone thus internalizes long-horizon capability directly into a small open model; we plan to release the complete training stack at https://github.com/AlibabaResearch/SignalCoverageRL.
Primary: Alibaba Research
All Institutions: Alibaba Research
The paper identifies two key failure modes in sparse-reward RL for LLM agents—signal starvation and policy drift—and proposes a minimalist protocol (CANOPY) that resolves them through scaled exploration and strict on-policy KL-anchored updates, achieving state-of-the-art results on AppWorld and SWE-bench with a small open model. This work provides a rigorous empirical and theoretical foundation for the viability of outcome-only RL in long-horizon interactive tasks, offering a practical and reproducible recipe that simplifies agent post-training by removing the need for dense rewards, SFT priors, or complex inference-time orchestration.
The paper proposes CANOPY, a minimalist RL protocol for training LLM agents on long-horizon tasks. The core contribution is a diagnostic framework identifying two failure modes in standard Group Relative Policy Optimization (GRPO) for sparse rewards: "signal starvation" (lack of gradient signal when rollout groups are homogeneous) and "policy drift" (entropy collapse due to unanchored optimization on small task pools). The solution involves scaling the rollout group size to ensure mixed outcomes (coverage), maintaining strict on-policy updates to prevent drift, and using a KL anchor to the base model. The methodology is sound, leveraging well-understood RL components (GRPO, KL penalty) but applying them with specific hyperparameter choices (large $n$, strict on-policy) that address the identified bottlenecks. The derivation of the advantage magnitude in degenerate groups provides a clear theoretical motivation for the increased exploration budget.
The experiments are extensive and compelling. The primary result is topping the AppWorld leaderboard with a Qwen3-14B model using only outcome-based RL, outperforming systems that rely on dense rewards, SFT priors, or complex orchestration. The ablation studies are rigorous, isolating the impact of group size, KL anchoring, and on-policy updates. The transfer to SWE-bench Verified demonstrates the generalizability of the principles, showing a 16.6 point improvement. The paper provides detailed training dynamics analysis, showing how entropy and coverage evolve, which supports the theoretical claims. The comparison against the base model and other RL variants is thorough.
The paper provides high reproducibility. It includes full configuration tables for both AppWorld and SWE-bench runs, details the hardware setup, and explains the fault quarantine mechanism for environment reliability. The authors explicitly state that they plan to release the complete training stack. The detailed appendix on the "w/o strict on-policy" variant and the discussion of compute costs (GPU-hours) are valuable for practitioners. The transparency regarding what was *not* logged (e.g., per-step quarantine counts) is a strength.
The primary limitation is the high compute cost; the main run required ~2,690 GPU-hours, which may be prohibitive for smaller labs. The paper acknowledges that the "minimalist" claim refers to architectural simplicity, not compute frugality. Additionally, the results are heavily dependent on the specific base models (Qwen3/3.5) and the AppWorld/SWE-bench environments; generalization to other domains or smaller models is not fully tested. The "signal starvation" diagnosis assumes independent rollouts, which may not hold perfectly in practice, though the empirical results support the approach.
This paper has significant potential to shift the paradigm in LLM agent training. By demonstrating that outcome-only RL can suffice for long-horizon tasks without dense rewards or complex scaffolding, it simplifies the post-training pipeline for open-source models. It challenges the prevailing belief that RL hits a ceiling on small models, suggesting that the ceiling is an artifact of training practice rather than a fundamental limit. This could lead to more efficient and scalable methods for training specialized agents in various domains. The paper identifies two key failure modes in sparse-reward RL for LLM agents—signal starvation and policy drift—and proposes a minimalist protocol (CANOPY) that resolves them through scaled exploration and strict on-policy KL-anchored updates, achieving state-of-the-art results on AppWorld and SWE-bench with a small open model. This work provides a rigorous empirical and theoretical foundation for the viability of outcome-only RL in long-horizon interactive tasks, offering a practical and reproducible recipe that simplifies agent post-training by removing the need for dense rewards, SFT priors, or complex inference-time orchestration.
AI tutors are most useful when they adapt to each student's strengths, weaknesses, and preferred guidance, but evidence about which guidance works for which student is sparse, slow, and costly to collect from real learners. Student simulators can provide this signal as a proxy, yet existing approaches are limited: state-tracking models fit student behavior but struggle to process explanations or corrections, while LLM role-play follows guidance fluently but does not reliably match the competence of the student being imitated. We present StudentSim, a training framework that turns sparse per-student data into individualized simulators through pooled training followed by per-student specialization. The resulting simulators both mirror a student's own responses and update them under tutor guidance. We also introduce StudentSimEval, a standardized protocol covering 60 students across chess, second-language English writing, and mathematics, using public learner datasets with de-identified records shared for research. StudentSimEval measures behavioral fidelity (F), or how well a simulator matches a student's responses, and guidance responsiveness (R), or how readily it updates under tutor guidance, with all methods fit and evaluated on the same records. Across all three domains, StudentSim outperforms GPT-5.4 on both metrics. In chess, StudentSim reaches F=0.51 and R=0.91, compared with 0.23 and 0.72 for GPT-5.4 and 0.45 and 0.27 for Maia2. As a proof of concept, using StudentSim as a reward model for tutor reinforcement learning produces a chess tutor that expert humans rate as more accurate, better-guided, and more personalized than a no-RL baseline and a tutor trained against a GPT-5.4 simulator reward. Code is available at https://github.com/microsoft/StudentSim.
Primary: Microsoft Research
All Institutions: Microsoft Research, University of Illinois Urbana-Champaign
StudentSim introduces a pooled-then-specialized training framework for LLM-based student simulators and a standardized evaluation protocol, demonstrating superior fidelity and responsiveness compared to general-purpose LLMs. The paper makes a strong contribution to educational AI by addressing the critical need for individualized student modeling, providing a rigorous benchmark, and validating the utility of these simulators in downstream tutor reinforcement learning tasks.
The paper proposes StudentSim, a framework for training LLM-based student simulators. The core methodological contribution is a two-stage training process: pooled training across multiple students to learn generalizable behavioral patterns, followed by per-student specialization to capture individual nuances. This addresses a specific gap where state-tracking models fail to process natural language guidance and generic LLM role-play fails to match specific student competence levels. The introduction of StudentSimEval, a standardized protocol with distinct metrics for behavioral fidelity (F) and guidance responsiveness (R), is a strong methodological addition that provides a rigorous way to evaluate these simulators.
The experiments are conducted across three diverse domains: chess, second-language English writing, and mathematics. The use of public, de-identified learner datasets enhances the validity of the results. The comparison against GPT-5.4 and Maia2 is significant, as GPT-5.4 represents a strong general-purpose baseline. The results showing StudentSim outperforming GPT-5.4 on both fidelity and responsiveness metrics are compelling. The proof-of-concept application using the simulator as a reward model for tutor RL, with positive human evaluation, further validates the utility of the proposed simulators.
The paper provides a link to the code repository (https://github.com/microsoft/StudentSim) and mentions the use of public datasets. The detailed description of the training framework and evaluation metrics in the appendices (implied by the section list) suggests good reproducibility. The availability of de-identified records for research further supports reproducibility efforts.
The primary limitation is the reliance on historical data for training, which may not capture the dynamic nature of real-time tutoring interactions. The generalization of the pooled training approach to unseen domains or student populations is not extensively explored. Additionally, the computational cost of training individualized simulators for each student could be a barrier to widespread adoption in resource-constrained settings.
This work has significant potential to improve the development of personalized AI tutors by providing a reliable way to simulate student behavior and response to guidance. The standardized evaluation protocol (StudentSimEval) could become a standard benchmark in the field of educational AI. The ability to use simulators for reward modeling in RL offers a new pathway for training more effective and personalized tutoring systems. StudentSim introduces a pooled-then-specialized training framework for LLM-based student simulators and a standardized evaluation protocol, demonstrating superior fidelity and responsiveness compared to general-purpose LLMs. The paper makes a strong contribution to educational AI by addressing the critical need for individualized student modeling, providing a rigorous benchmark, and validating the utility of these simulators in downstream tutor reinforcement learning tasks.
Scaling Transformers has driven large gains in language modeling, but transplanting this to behavior-sequence modeling in production ranking is challenging: recommendation differs in signal quality, where behavior sequences are noisy, temporally irregular, and sparsely supervised, and in computation asymmetry, where each request scores many candidates against one shared user history under tight latency budgets. We propose ReST, a recommendation-native Transformer scaling framework. For signal quality, it introduces a sequence encoder with dual-gated attention, rotary positional and temporal embedding, stabilized residual normalization, and training-only auxiliary objectives. For computation asymmetry, it factorizes ranking into a heavy reusable encoder and a lightweight cross decoder with projection-free KV attention and token-specific parameterization, coupling user-level shared-prefix training with shared-prefix serving for compute-once, decode-many-times ranking. Across industrial and public benchmarks, ReST achieves higher accuracy and scales more consistently along sequence length, depth, and width, where LLM-style Transformer blocks saturate. A one-week online A/B test on a production advertising platform improves online AUC by 1.31% and lifts a core revenue metric by 11.93% within a 50 ms P99 budget; ReST has since been fully deployed in production, showing that behavior-sequence scaling remains a promising, under-exploited axis for production ranking.
Primary: ByteDance
All Institutions: ByteDance
ReST introduces a recommendation-native Transformer scaling framework that addresses signal quality and computation asymmetry in industrial ranking. By factorizing the model into a reusable encoder and lightweight decoder and introducing specific denoising and temporal mechanisms, it achieves consistent scaling gains and significant online business impact, demonstrating that behavior-sequence scaling is a viable axis for production ranking.
The paper proposes ReST, a framework that adapts Transformer scaling laws to industrial recommendation ranking by addressing two specific mismatches with LLMs: signal quality (noisy, sparse supervision) and computation asymmetry (one user history vs. many candidates). The methodology introduces "Rec-Native" components: Dual-Gated Attention (DGA) to filter noisy behavior tokens, Rotary Temporal Embedding (RoTE) to handle irregular time gaps, and Stabilized Residual Normalization (SRN) to prevent depth collapse under sparse supervision. Crucially, it factorizes the model into a heavy, reusable sequence encoder and a lightweight cross-decoder, enabling "compute-once, decode-many-times" serving. The introduction of auxiliary objectives to combat "sequence starvation" (where the DLRM branch shortcuts the gradient flow to the sequence branch) is a significant practical insight.
The experimental validation is strong, featuring a large-scale industrial dataset from TikTok Shop Ads and public benchmarks (MovieLens, Amazon-Books). The paper demonstrates consistent scaling benefits in length, depth, and width where LLM-style baselines saturate. The most compelling evidence is the online A/B test reporting a 1.31% AUC improvement and an 11.93% lift in core revenue metrics within a strict 50ms P99 latency budget. The ablation studies effectively isolate the contributions of DGA, RoTE, and the auxiliary losses.
Reproducibility is moderate. While the paper provides detailed architectural descriptions and hyperparameters for public benchmarks, the industrial dataset is proprietary. The specific implementation of the "shared-prefix serving" infrastructure is described conceptually but not open-sourced. The public benchmark results are reproducible given the RecBole framework details, but the primary claims rely on the industrial deployment which cannot be independently verified.
The primary limitation is the lack of open-source code and the proprietary nature of the main evaluation dataset. The "sequence starvation" phenomenon is specific to hybrid DLRM+Transformer architectures; its applicability to pure end-to-end recommendation models is less clear. Additionally, the scaling laws fitted are descriptive over a limited compute range and may not hold for orders of magnitude larger scales.
This work bridges the gap between LLM scaling research and industrial recommendation systems, providing a blueprint for efficiently scaling sequence models in latency-constrained environments. The insights on computation asymmetry and auxiliary supervision are likely to influence future architectures in ad-tech and e-commerce. ReST introduces a recommendation-native Transformer scaling framework that addresses signal quality and computation asymmetry in industrial ranking. By factorizing the model into a reusable encoder and lightweight decoder and introducing specific denoising and temporal mechanisms, it achieves consistent scaling gains and significant online business impact, demonstrating that behavior-sequence scaling is a viable axis for production ranking.
Sampling from distributions conditioned on desired semantic properties is an emerging challenge in modern generative modeling. Metropolis-Hastings (MH) provides a principled route to conditional sampling, but requires access to exact pointwise target-density evaluations, which are not available in generative settings. Meanwhile, pairwise comparisons by humans or model "judge" are highly accessible and have proved valuable across diverse applications. We introduce Pref-MH, a general exact MH sampler for judge-induced conditional distributions using only stochastic binary pairwise comparisons. Our key observation is that the MH unnormalized density ratio matches the preference odds of the Bradley-Terry (BT) choice model. The central challenge is that while MH requires precise ratio computation, BT judges provide only sampled binary feedback. To this end, we develop a valid accept/reject rule whose resulting Markov chain provably converges to the target distribution. We further show that, for a fixed proposal kernel and budget, Pref-MH is optimal in the Peskun-Tierney sense among this class of exact reversible acceptance rules. Experiments on text generation and molecular design with LLM judges, as well as image generation with VLM judges, demonstrate that Pref-MH provides a practical and flexible approach to conditional sampling when comparative feedback is relatively easy to obtain.
Primary: Technion -- Israel Institute of Technology
All Institutions: Technion -- Israel Institute of Technology
[One sentence main contribution]. The paper presents Pref-MH, an exact MCMC sampler that utilizes only stochastic binary pairwise comparisons to sample from judge-induced conditional distributions, proving its validity and optimality within a fixed budget constraint. [Comprehensive analysis of the technical contribution, methodology, and significance to the field]. The technical contribution is significant as it bridges the gap between the theoretical requirements of MCMC (exact density ratios) and the practical availability of comparative feedback in modern generative AI. By proving that plug-in estimators fail and providing an optimal exact rule, the authors establish a robust foundation for preference-based sampling. The experiments demonstrate practical utility across diverse tasks, reinforcing the method's versatility. The work is well-positioned to influence future research on inference-time alignment and conditional generation.
The paper introduces Pref-MH, a Metropolis-Hastings sampler that operates using only stochastic binary pairwise comparisons (Bradley-Terry model) rather than pointwise density evaluations. The core theoretical contribution is the derivation of an exact accept/reject rule that preserves the target stationary distribution despite the noise in the judge's feedback. The authors prove an impossibility result for plug-in estimators and demonstrate that their proposed $N$-vote rule is optimal in the Peskun-Tierney sense among exact reversible rules with a fixed budget. The method is modular, supporting multiple judges for different properties.
Experiments cover synthetic validation, text generation (Llama 3.1), image generation (SDXL-Turbo), and molecular design (Qwen 235B). The synthetic experiment rigorously validates the theoretical claims by showing convergence to the true target while plug-in baselines fail. The molecular design experiment shows improvements in MolSkill scores compared to MARS and pointwise MH baselines. The text and image experiments demonstrate the flexibility of the method across modalities.
The code is available on GitHub. The paper provides detailed algorithmic descriptions and hyperparameters for the experiments. The use of specific LLMs and diffusion models makes the experiments somewhat resource-intensive but reproducible for labs with access to such models.
The method relies on the Bradley-Terry assumption for the judge's preferences, which may not hold perfectly for complex LLM judges. The computational cost is high due to the need for multiple judge queries per MCMC step. The optimality is only within the class of exact reversible rules with fixed budgets, not necessarily globally optimal compared to approximate methods.
This work provides a principled framework for using comparative feedback in generative modeling, which is highly relevant given the prevalence of LLM-as-a-judge paradigms. It offers a rigorous alternative to heuristic reward-based sampling methods, potentially improving the reliability of conditional generation in high-stakes domains like drug discovery. [One sentence main contribution]. The paper presents Pref-MH, an exact MCMC sampler that utilizes only stochastic binary pairwise comparisons to sample from judge-induced conditional distributions, proving its validity and optimality within a fixed budget constraint. [Comprehensive analysis of the technical contribution, methodology, and significance to the field]. The technical contribution is significant as it bridges the gap between the theoretical requirements of MCMC (exact density ratios) and the practical availability of comparative feedback in modern generative AI. By proving that plug-in estimators fail and providing an optimal exact rule, the authors establish a robust foundation for preference-based sampling. The experiments demonstrate practical utility across diverse tasks, reinforcing the method's versatility. The work is well-positioned to influence future research on inference-time alignment and conditional generation.
Research planning is the decisive capability of AI scientists. Yet a research plan admits no verifiable answer, so reinforcement learning lacks the environment it requires: tasks paired with a critic. Rubrics extracted from scientific papers can supply the critic. Existing pipelines, however, draw the question and the criteria from the same content, so the reward can be earned by paraphrase. The rubric is further compressed into a single scalar per rollout. We introduce PaperGym, a unified framework that turns each research paper into a complete training environment. PaperGym exploits the structure of a paper: the question is synthesized from the research goal and background, while the criteria are derived from the method and experiments. The criteria span methodological innovation and experimental design, and criterion leakage falls to 3.7%, versus 11.90% to 34.10% in existing datasets. Training uses the rubric twice: first as privileged context for OPSD's self-teacher, then as the reward for GRPO. Across Qwen3-1.7B/4B/8B, this schedule outperforms supervised fine-tuning, either stage alone, and the reverse ordering, improving five-benchmark averages by +5.6, +5.0, and +4.8 points. With the recipe held fixed, models trained on PaperGym-20k win 58.1% of three-way comparisons, against 28.2% for RubricHub Science. The trained Qwen3-8B reaches 73.48 on ResearchQA, above the far larger Kimi K2.6. We release the pipeline, the 20,000-instance corpus PaperGym-20k, and the benchmarks PaperGym-Innov and PaperGym-Design.
Primary: Zhejiang University
All Institutions: Zhejiang University
The paper introduces a robust framework for generating research plans by decoupling question and answer sources to reduce leakage and employing a two-stage OPSD+GRPO training strategy. It demonstrates significant improvements over baselines and competitive performance against larger models, offering a reproducible pipeline and dataset for the scientific AI community.
The paper proposes "PaperGym," a framework that converts research papers into training environments for research-plan generation. The core methodological contribution is a structured data pipeline that decouples the generation of the research question (from Goal/Background) from the reference answer (from Method/Experiments) to minimize criterion leakage. It introduces a two-stage training paradigm: first, On-Policy Self-Distillation (OPSD) using the rubric as privileged context to establish a broad prior, followed by Group Relative Policy Optimization (GRPO) using the rubric as a reward signal. The use of rubrics as both a distillation teacher context and a reinforcement learning reward is a clever integration of dense and sparse supervision signals. The method addresses the "cold start" instability of GRPO by using OPSD to structure the initial policy distribution, a hypothesis supported by entropy dynamics analysis.
The experiments are extensive, covering three model scales (Qwen3-1.7B/4B/8B) and multiple benchmarks (in-domain and out-of-domain). The paper demonstrates that the two-stage approach outperforms SFT, single-stage OPSD, single-stage GRPO, and the reverse ordering (GRPO+OPSD). The win-rate analysis against RubricHub Science and the base model provides strong evidence of the data quality. The comparison with larger general-purpose models (Kimi K2.6, GPT-5.1) shows competitive performance for a small model, which is a significant result. The ablation studies on rubric sources and model scales are thorough.
The paper releases the pipeline, the 20,000-instance corpus (PaperGym-20k), and benchmarks. Detailed hyperparameters, prompts, and training configurations are provided in the appendix. The use of standard frameworks (verl, vLLM) and open-source base models (Qwen3) enhances reproducibility. The specific scoring protocols and judge prompts are fully disclosed.
The primary limitation is the reliance on LLM-as-a-judge for both data generation (rubric creation) and evaluation, which introduces potential biases and costs. The "criterion leakage" metric, while improved, is still non-zero (3.7%), and the effectiveness of the rubric as a reward signal depends heavily on the quality of the LLM judge. The paper focuses on Qwen3 models; generalization to other architectures is not tested. The computational cost of the two-stage training and the multi-call rubric evaluation during GRPO is significant.
This work provides a scalable method for generating high-quality training data for scientific reasoning tasks, which is a critical bottleneck in AI-for-Science. The framework could be adapted to other domains where expert feedback is scarce but structured documents exist. The two-stage OPSD+GRPO recipe may be applicable to other tasks requiring dense supervision and outcome-based optimization. The paper introduces a robust framework for generating research plans by decoupling question and answer sources to reduce leakage and employing a two-stage OPSD+GRPO training strategy. It demonstrates significant improvements over baselines and competitive performance against larger models, offering a reproducible pipeline and dataset for the scientific AI community.
Multimodal models often build on architectures designed for generative vision-language modeling, typically combining separately pretrained vision encoders with causal language models. Visual document retrievers such as ColPali repurpose these models as encoders, carrying over the parameter and compute overhead of a VLM for a non-generative task. We introduce NeoMME, a family of 260M and 800M-parameter Multimodal and Multilingual bidirectional Encoders that process multilingual text and raw image patches in a single bidirectional Transformer encoder. Both models are pretrained from scratch with a masked discrete-diffusion text objective, conditioned on visible image patches for multimodal examples. Both support a 16,384-token context, enough to encode up to two standard 4K UHD images. To demonstrate its downstream capabilities, we fine-tune NeoMME with jointly trained dense and late-interaction heads. On the ViDoRe v3 benchmark, the resulting NeoMME-Retriever 260M outperforms all evaluated models strictly below 800M parameters with 0.523 nDCG@10, while NeoMME-Retriever 800M reaches 0.556. At a matched 2048x2048 image input size on an NVIDIA L40S, NeoMME-260M encodes pages with about 2x the throughput of ColModernVBERT. Hierarchical token pooling and asymmetric quantization compress late-interaction multimodal document embeddings by 255x while preserving over 95% of baseline nDCG@10. We contribute NeoMME to Hugging Face Transformers and release the pretrained backbone and retrieval-compatible checkpoints under Apache 2.0 at https://hf.co/collections/Hcompany/neomme.
Primary: H Company
All Institutions: H Company
NeoMME introduces a lightweight, single-tower multimodal encoder trained with discrete diffusion that achieves competitive retrieval performance with significantly higher efficiency than repurposed VLMs. The paper makes a solid contribution to the field of efficient multimodal learning by demonstrating that high-quality document retrieval can be achieved without the computational burden of generative architectures, offering a practical and open-source alternative for industry and research applications.
The paper proposes NeoMME, a single-tower, bidirectional multimodal encoder that processes raw image patches and multilingual text tokens jointly without a separate vision tower or causal language model decoder. The architecture is trained from scratch using a masked discrete-diffusion objective, which is a novel training paradigm for encoder-only multimodal models. The design aims to eliminate the parameter and compute overhead associated with repurposing generative Vision-Language Models (VLMs) for non-generative retrieval tasks. The inclusion of dense and late-interaction heads for retrieval, along with hierarchical token pooling and asymmetric quantization for compression, represents a comprehensive engineering and methodological contribution to efficient multimodal retrieval.
The experiments focus on the ViDoRe v3 benchmark, a standard for visual document retrieval. The 260M parameter model outperforms all models strictly below 800M parameters, achieving 0.523 nDCG@10, while the 800M model reaches 0.556. Throughput comparisons against ColModernVBERT on an NVIDIA L40S show a 2x improvement for the 260M model at 2048x2048 resolution. The compression results are significant, claiming a 255x reduction in embedding size with >95% retention of nDCG@10. However, the evaluation is limited to a single benchmark suite (ViDoRe) and lacks comparisons against other state-of-the-art encoder-only models or broader multimodal understanding tasks.
The paper claims to release the pretrained backbone and retrieval-compatible checkpoints under Apache 2.0 on Hugging Face, which is a strong positive for reproducibility. The architecture details (single-tower, bidirectional, 16k context) are clearly defined. However, the specific hyperparameters for the discrete-diffusion pretraining and the exact implementation details of the quantization scheme are not fully detailed in the provided abstract/summary, though the code release mitigates this.
The primary limitation is the narrow scope of evaluation, focusing almost exclusively on document retrieval (ViDoRe). It is unclear how well the encoder generalizes to other multimodal tasks like visual question answering or image captioning. The "H Company" affiliation is opaque, making it difficult to assess the institutional prestige or long-term maintenance of the project. The lack of ablation studies on the discrete-diffusion objective versus standard masked language modeling is a gap.
This work addresses a critical inefficiency in current multimodal pipelines: the use of heavy generative models for simple encoding tasks. By providing a lightweight, efficient, and open-source encoder, it could significantly lower the barrier to entry for multimodal retrieval systems, enabling deployment on edge devices or in high-throughput server environments. The focus on multilingual support also broadens its applicability globally. NeoMME introduces a lightweight, single-tower multimodal encoder trained with discrete diffusion that achieves competitive retrieval performance with significantly higher efficiency than repurposed VLMs. The paper makes a solid contribution to the field of efficient multimodal learning by demonstrating that high-quality document retrieval can be achieved without the computational burden of generative architectures, offering a practical and open-source alternative for industry and research applications.
We present Qwen-Drive-1.0, an initial step towards a vision-language foundation model for autonomous driving. Qwen-Drive-1.0 retains the architecture of the pretrained vision-language model (VLM) and integrates 3D perception, visual question answering, and motion planning within a unified framework. An external bird's-eye-view (BEV) perception head jointly performs 3D object detection, semantic occupancy prediction, and BEV map segmentation. It serves as a probe of the 3D information accessible from the shared representations and provides an explicit, inspectable interface to 3D scene structure. A Planning Expert conditions on shared VLM representations to generate future ego trajectories. A staged training recipe combines driving supervision with general-purpose vision-language data to acquire driving-specific competence while helping preserve broad visual understanding and instruction-following capabilities. Experiments demonstrate strong 3D perception and driving scene understanding while largely preserving general vision-language capability. Comprehensive evaluations across open-loop, pseudo-closed-loop, and closed-loop settings further show highly competitive motion-planning performance.
Primary: Qwen Team (Alibaba Group)
All Institutions: Huazhong University of Science and Technology, Qwen Team
Qwen-Drive-1.0 introduces a unified vision-language foundation model for autonomous driving that integrates 3D perception, VQA, and motion planning without altering the pretrained VLM architecture. The paper demonstrates that by attaching external BEV perception and planning experts and employing a staged training recipe with reinforcement learning, it is possible to achieve competitive 3D perception and motion planning performance while preserving general visual-language capabilities, addressing key limitations of previous domain-adapted VLMs.
The paper proposes Qwen-Drive-1.0, a unified framework that integrates 3D perception, visual question answering (VQA), and motion planning within a single pretrained Vision-Language Model (VLM) architecture. The core methodological contribution is the attachment of two external modules to a frozen or lightly tuned VLM backbone (Qwen3.5-4B): a BEV perception head for explicit 3D scene understanding (detection, occupancy, map segmentation) and a Planning Expert for trajectory generation. The BEV head uses a depth-based view transform to lift 2D features into 3D volumes, leveraging both low-level vision encoder features and high-level VLM semantic features. The Planning Expert employs flow matching with a diffusion transformer, conditioning on cached VLM keys/values to generate trajectories. A key methodological strength is the staged training recipe: (1) pretraining the perception head, (2) joint fine-tuning of perception and VQA to adapt the shared representations, (3) pretraining the planning expert, and (4) reinforcement learning (RL) for planning. The RL stage is particularly interesting, using a stochastic policy gradient approach with low-frequency temporal subspace perturbations to ensure smooth trajectory diversity, optimizing task-level rewards like collision avoidance and drivable area compliance. The data pipeline is robust, unifying heterogeneous datasets (nuScenes, OpenScene, NAVSIM, Waymo) through label alignment and spatial normalization.
The evaluation is comprehensive, covering 3D perception (nuScenes, OpenScene), driving VQA, general VQA, and motion planning (NAVSIM, WOD-E2E, AlpaSim). The model achieves competitive 3D perception metrics (43.95 mAP on nuScenes) while significantly outperforming the base VLM on driving-specific tasks. Crucially, it preserves general VQA capabilities, addressing the catastrophic forgetting issue common in domain adaptation. The motion planning results are strong, with a PDMS of 90.7 on NAVSIM and a Rater Feedback Score of 7.91 on WOD-E2E. The ablation studies effectively demonstrate the necessity of joint training (Stage 2) for 3D perception performance, showing that head-only training is insufficient. The comparison with dedicated detectors (BEVFormerV2, PETR) highlights the trade-off between unified representation and specialized performance, though the unified model remains highly competitive.
The paper provides detailed descriptions of the architecture, training stages, and data processing pipelines. Specific hyperparameters, such as learning rates, batch sizes, and flow matching parameters, are included. The use of public datasets (nuScenes, OpenScene, NAVSIM, Waymo) and open-source VLMs (Qwen3.5) enhances reproducibility. However, the specific implementation details of the "Qwen3.5-Plus" and "Qwen3.7-Plus" models used for data generation and auditing are proprietary, which may limit full reproduction of the data curation process. The code and model weights are not explicitly linked in the provided text, but the use of standard components suggests high reproducibility if released.
The model relies on a 4B parameter VLM, which may limit its reasoning capabilities compared to larger models. The RL stage, while effective, introduces complexity in reward design and stochastic policy optimization. The paper acknowledges that textual VQA targets do not directly constrain 3D layout, though the BEV head mitigates this. The cross-dataset transfer for occupancy prediction shows some negative transfer, indicating challenges in unifying diverse annotation styles. The computational cost of training a unified model with multiple heads and RL stages is likely high, though inference efficiency is not deeply analyzed.
This paper represents a significant step towards integrated autonomous driving systems that can handle both perception and high-level reasoning. By preserving general VLM capabilities, it enables a single model to serve both driving and cockpit functions, reducing hardware costs. The approach of using explicit 3D perception heads as probes for VLM representations offers a practical way to evaluate and enhance spatial understanding in LLMs. The staged training recipe and data unification strategies provide a blueprint for building robust, multi-task driving models. This work could influence future research in VLA (Vision-Language-Action) models, encouraging the integration of explicit geometric reasoning with language-based planning. Qwen-Drive-1.0 introduces a unified vision-language foundation model for autonomous driving that integrates 3D perception, VQA, and motion planning without altering the pretrained VLM architecture. The paper demonstrates that by attaching external BEV perception and planning experts and employing a staged training recipe with reinforcement learning, it is possible to achieve competitive 3D perception and motion planning performance while preserving general visual-language capabilities, addressing key limitations of previous domain-adapted VLMs.
Much of human health and function unfolds beyond the clinic, through the movements of everyday life. Wrist-worn accelerometers capture these movements continuously, yet their rich signals are often reduced to a small set of predefined behavioural summary measures. Here, we present Sensori, a self-supervised foundation model that learns general-purpose health representations directly from 24 hours of raw tri-axial wrist movement. We developed and evaluated the model across four population-based cohorts from the United Kingdom, China and the United States, comprising 122,640 participants contributing 683,617 person-days of free-living recordings. Sensori condensed each day of movement into a representation that captured diverse movement behaviours, demographic characteristics, health axes and physical function. Evaluation in independent cohorts showed that these representations generalised across populations and measurement settings without retraining. When added to common clinical covariates, Sensori significantly improved prevalent disease classification for 52 of 102 eligible conditions (median delta AUROC, 0.060; range, 0.012-0.242) and incident disease risk prediction for 26 of 87 eligible conditions (median delta Uno's C-index, 0.064; range, 0.025-0.172), with the largest gains for neurological and psychiatric disorders. These findings establish 24-hour wrist movement as a rich and scalable source of health information, with the potential to support passive health monitoring and disease prediction at population scale.
Primary: University of Oxford
All Institutions: University of Oxford, University College London, University of Birmingham, Peking University, National and Kapodistrian University of Athens, University Hospitals Birmingham NHS Foundation Trust, NIHR Biomedical Research Centre, Wellcome Trust, Big Data Institute
The paper presents a highly significant contribution to health informatics and self-supervised learning by introducing Sensori, a foundation model that learns general-purpose health representations from 24-hour wrist movement data at an unprecedented scale (122k participants, 683k person-days). The rigorous evaluation across four international cohorts demonstrates strong cross-population generalization and clinically meaningful improvements in disease prediction, particularly for neurological and psychiatric conditions, establishing raw wrist acceleration as a rich, scalable source of health information.
The paper introduces Sensori, a self-supervised foundation model for 24-hour wrist-worn accelerometer data. The architecture is a hybrid of a 1D convolutional encoder (adapted from wav2vec 2.0) and a Transformer backbone. The key methodological contribution is the dual pretraining objective: masked reconstruction at the 5-minute embedding level to capture intra-day temporal dependencies, and participant-level contrastive learning to capture stable inter-day individual characteristics. This approach effectively handles the long-sequence nature of 24-hour data (864,000 samples at 10Hz) by progressively pooling to 5-minute embeddings before applying the Transformer. The use of linear probing for evaluation is standard and appropriate for assessing representation quality.
The experimental scale is exceptional, utilizing 122,640 participants and 683,617 person-days across four major population cohorts (UK Biobank, China Kadoorie Biobank, ELSA, NHANES). The evaluation is comprehensive, covering activity recognition benchmarks, prediction of demographic/health traits, and disease risk prediction. The results show strong generalization across cohorts without retraining. The improvement in disease prediction (median delta AUROC 0.060) is statistically significant and clinically relevant, particularly for neurological and psychiatric disorders. The comparison against conventional handcrafted features and other foundation models (MOMENT, Chronos-2) is rigorous.
High. The authors provide code, pretrained weights, and a project website. The data sources are well-defined, though access to UKB and CKB requires approval. The preprocessing steps (actipy) are open-source. The model architecture and hyperparameters are detailed in the methods section.
The model is trained primarily on older populations (median age >60 in most cohorts), limiting generalizability to younger demographics. Disease analysis is restricted to UKB due to data availability. The paper acknowledges that improvements in incident disease risk may reflect prodromal states rather than true predictive risk. The computational cost of processing 24-hour sequences is non-trivial, though the 40M parameter model is manageable.
This work has significant potential for passive health monitoring and digital biomarker discovery. By learning general-purpose representations from raw movement data, it enables scalable screening for a wide range of conditions without condition-specific model training. It bridges the gap between raw sensor data and clinical utility, offering a pathway for integrating wearable data into routine healthcare. The paper presents a highly significant contribution to health informatics and self-supervised learning by introducing Sensori, a foundation model that learns general-purpose health representations from 24-hour wrist movement data at an unprecedented scale (122k participants, 683k person-days). The rigorous evaluation across four international cohorts demonstrates strong cross-population generalization and clinically meaningful improvements in disease prediction, particularly for neurological and psychiatric conditions, establishing raw wrist acceleration as a rich, scalable source of health information.
Protein structure modeling rests on a single computational primitive: the interaction between what a residue is (sequence content) and where it sits (three-dimensional geometry). What is the expressive limit of this layer class? We show that the complete bilinear operator over content-geometry outer products--the sufficient statistic of all second-order interactions--is the expressive ceiling, while the additive message passing of mainstream geometric GNNs is provably blind to content-geometry binding. We then introduce Hyper-Fold, a rank-K separable convolutional backbone approaching this ceiling at message-passing cost: each radius neighborhood is organized into a sequence hyperedge and a contact hyperedge, modulated by an edge-conditioned matrix-valued operator factorized into K learned basis operators with geometry-generated coefficients. Across enzyme function prediction, fold classification, and ligand binding site detection, Hyper-Fold and its hierarchical variant Hyper-Fold-Deep achieve the best results among protein-specific structure encoders; Hyper-Fold-Pocket, an anchored set-prediction head, surpasses UniSite-3D on UniSite-DS and two zero-shot benchmarks with no sequence language model features, 68x fewer parameters, and 4.8x lower latency--suggesting that a sufficiently expressive 3D backbone recovers information that fusion architectures previously borrowed from evolution-scale pretraining.
Primary: Shanghai University
All Institutions: Shanghai University, Tsinghua University, Xi'an Jiaotong University, Zhejiang University
The paper establishes a theoretical expressive limit for sequence-geometry layers in protein modeling and introduces Hyper-Fold, a rank-K hypergraph-based architecture that approaches this limit with message-passing efficiency, achieving state-of-the-art results in pocket detection and function prediction without relying on sequence language models.
The paper proposes a theoretically grounded framework for protein structure modeling, defining an "expressive ladder" for sequence-geometry interaction layers. The core theoretical contribution is the proof that additive message passing (used in GearNet, GVP, etc.) is provably blind to content-geometry binding, while the complete bilinear operator over outer products represents the expressive ceiling. To make this ceiling computationally feasible, the authors introduce Hyper-Fold, a rank-K separable convolutional backbone that approximates the bilinear operator using hypergraph structures (sequence and contact hyperedges). The method is mathematically sound, leveraging factorization to reduce complexity from $O(C^2 d_g)$ to $O(K C C')$, which is comparable to standard message passing. The design of the "Hyper-Fold-Pocket" head, which uses structure-anchored set prediction without sequence language model features, is a clever architectural choice that directly tests the hypothesis that expressive 3D backbones can replace evolutionary information.
The experimental evaluation is rigorous and comprehensive. The authors benchmark on three distinct tasks: enzyme function prediction (EC), fold classification, and ligand binding site detection (Pocket). They demonstrate state-of-the-art or near-state-of-the-art results on all three. Notably, Hyper-Fold-Pocket outperforms UniSite-3D (which uses ESM features) on zero-shot benchmarks (HOLO4K, COACH420) without using any sequence language model features, achieving this with 68x fewer parameters and 4.8x lower latency. The ablation studies effectively isolate the contribution of the rank-K kernel, the hyperedge structure, and the contrastive denoising training strategy. The comparison against the "expressive ladder" rungs (additive vs. scalar vs. channel vs. matrix gating) provides strong empirical evidence supporting the theoretical claims.
The paper includes a detailed reproducibility statement, specifying architecture configurations, training recipes, and providing anonymized source code and weights. The use of standard benchmarks (UniSite-DS, HOLO4K, COACH420) and clear evaluation metrics ([email protected]/0.5, Fmax) facilitates independent verification. The complexity analysis and latency measurements are clearly defined (A100-80GB, batch size 1, synthetic length-300 protein).
The primary limitation is the reliance on radius graphs, which may not capture long-range interactions as effectively as attention-based methods for very large proteins, although the paper argues for the efficiency of the linear scaling. The theoretical "ceiling" is defined for second-order interactions; it is unclear how this extends to higher-order geometric constraints. Additionally, while the model is efficient, the hypergraph construction adds implementation complexity compared to standard GNNs. The zero-shot generalization, while impressive, is tested on a limited set of benchmarks.
This work has significant implications for the field of geometric deep learning in biology. By proving the limitations of additive message passing and providing a scalable alternative, it offers a new direction for designing protein structure encoders. The finding that expressive 3D backbones can substitute for sequence language model features in specific tasks (like pocket detection) challenges the prevailing paradigm of hybrid sequence-structure models and could lead to more efficient, structure-only pipelines for drug discovery and protein engineering. The paper establishes a theoretical expressive limit for sequence-geometry layers in protein modeling and introduces Hyper-Fold, a rank-K hypergraph-based architecture that approaches this limit with message-passing efficiency, achieving state-of-the-art results in pocket detection and function prediction without relying on sequence language models.
Tool-using agents are commonly evaluated by a single bit: whether an end-to-end workflow completed. This metric fails to distinguish failures that matter in production, such as expired credentials, malformed payloads, or correct execution followed by incorrect final delivery. We introduce APIFlow-Bench, a fully auditable benchmark for long-horizon, dependent REST-API workflows that decomposes performance into seven engineering capabilities and requires agents to produce answers supported by the actual call path. We generate synthetic API worlds forward, subtask by subtask; each subtask is admitted only after a zero-LLM self-test triad verifies its grader and an oracle establishes solvability, and an adversarial audit identified and fixed six grader exploits. Grading is deterministic and provenance-sensitive: a state check traces a mock-minted canary through the API data flow to the response the answer must originate from, and a typed answer card is verified field by field. We release all answer keys and 44,362 unredacted execution transcripts. Across 19 frontier and open-weight models under one neutral scaffold, we find: (1) longer dependency chains degrade success, from 93% on individual subtasks to 74% on clean 20-subtask chains and 61% when including the 8% of chain trials that a model-consensus screen flags as passed by no model; (2) reliability separates models more than best-case capability, with best-of-five spanning seven points but all-five-of-five reliability spanning 44 points; (3) the independent-error account of compounding failure does not fit the data: pass rates on 20-subtask chains are 33 percentage points above the product of subtask-level rates, and on the clean slice 77% of failing runs reached the correct final state and failed only at delivery.
Primary: Postman
All Institutions: Postman
APIFlow-Bench introduces a rigorous, provenance-gated benchmark for long-horizon API workflows that reveals a critical gap between agent capability and reliability, showing that final delivery failures dominate over state execution errors in long dependent chains. The paper's contribution lies in its methodological rigor in benchmark generation and validation, providing a reusable framework for creating auditable synthetic benchmarks, and its empirical findings that challenge the independent-error model of compounding failure, offering actionable insights for improving agent robustness in production-like environments.
The paper introduces a rigorous methodology for generating and validating synthetic API benchmarks. The core innovation is the "zero-LLM self-test triad" and the provenance-gated grading system. By using mock-minted canaries to trace data flow, the authors ensure that agents cannot pass by guessing or memorizing answers; they must execute the correct API calls. The validation stack (oracle solvability, golden replay, adversarial audit) is a significant methodological contribution to the field of generated benchmarks, addressing the common issue of grader exploits and unsolvable tasks. The decomposition of performance into seven specific engineering capabilities (authentication, discovery, schema repair, etc.) provides a granular view of agent competence that binary success metrics miss.
The evaluation is extensive, covering 19 frontier and open-weight models across 467 tasks. The key empirical finding is the "level collapse" with chain length, showing that while individual subtasks are easy (93% pass), long dependent chains degrade performance significantly (74% pass). The paper provides a strong analysis of failure modes, distinguishing between state execution failures and final delivery failures. The finding that 77% of failing runs reached the correct final state but failed at delivery is a surprising and valuable insight for practitioners. The reliability analysis (pass@5 vs pass^5) correctly identifies that consistency, not just capability, is the primary differentiator between models in long-horizon tasks.
High. The authors release the full harness, frozen task bank, answer keys, and 44,362 unredacted execution transcripts. The use of content-hash-pinned manifests and deterministic evaluators ensures that results are reproducible. The paper provides clear instructions for reproducing the leaderboard and verifying the bank integrity. The open-sourcing of the transcripts allows for independent analysis of failure modes, which is a best practice in benchmarking.
The primary limitation is the small number of sampling units (11 full-length worlds) for the headline chain-20 slice, leading to wide confidence intervals and overlapping model rankings. The paper acknowledges that the "clean" slice is heavily influenced by two specific world families, which may limit the generalizability of the failure mode analysis. Additionally, the benchmark is REST-only, and the synthetic nature of the API worlds, while controlled, may not fully capture the complexity and unpredictability of real-world production APIs. The entanglement of the generator family with the oracle and reviser models is a potential confounding factor, though mitigated by deterministic validation.
This paper has high potential impact on the development of tool-using agents. By providing a benchmark that distinguishes between capability and reliability, and by exposing specific failure modes like final delivery errors, it guides developers toward more robust agent designs. The provenance-gated grading approach is a reusable technique for other benchmark domains. The findings on the non-multiplicative nature of long-horizon failure challenge common assumptions about compounding errors and suggest that agents fail due to exposure to specific failure surfaces rather than hidden information accumulation. This insight is valuable for designing better error recovery and state management mechanisms in agents. APIFlow-Bench introduces a rigorous, provenance-gated benchmark for long-horizon API workflows that reveals a critical gap between agent capability and reliability, showing that final delivery failures dominate over state execution errors in long dependent chains. The paper's contribution lies in its methodological rigor in benchmark generation and validation, providing a reusable framework for creating auditable synthetic benchmarks, and its empirical findings that challenge the independent-error model of compounding failure, offering actionable insights for improving agent robustness in production-like environments.
Latent generative models typically follow a two-stage pipeline, training a variational autoencoder for reconstruction and then a generative model on the frozen latent space. Since reconstruction-optimized latents are not necessarily generation-friendly, jointly training both models is an appealing alternative. However, direct end-to-end training remains challenging, as it is prone to latent collapse and faces a generation-reconstruction conflict. We revisit this problem by analyzing how different objectives shape the latent space and identify two key insights. First, the entropy term in the Kullback-Leibler divergence objective is essential for preventing collapse: reconstruction and prior fitting tend to shrink the posterior, while entropy preserves non-degenerate latent uncertainty. Second, reconstruction and generation exhibit asymmetric learning dynamics: reconstruction is fast and strongly supervised, whereas generation is slower and harder to optimize. Based on these insights, we achieve the first direct end-to-end training without latent collapse and propose GenFirst, a simple generation-before-reconstruction strategy. The generative objective first shapes the latent space under weak reconstruction pressure, after which reconstruction is progressively strengthened to recover visual details. We validate GenFirst with continuous autoregressive priors with exact likelihoods and SiT priors with implicit likelihoods. With our end-to-end objective and GenFirst, SiT achieves a gFID of 0.97 with CFG and 1.45 without CFG on ImageNet-256, while MMDiT reaches a GenEval score of 0.90 on text-to-image generation. Beyond image generation, we extend the framework to shared visual latents for generation and representation learning, and to continuous unified text-image generation. These results demonstrate the generality of stable end-to-end latent learning across generative priors and modalities.
Primary: University of Science and Technology of China
All Institutions: University of Science and Technology of China
The paper identifies the root cause of latent collapse in end-to-end training (prior-entropy imbalance) and proposes a simple, effective strategy (GenFirst) to resolve the generation-reconstruction conflict, achieving state-of-the-art results on ImageNet and text-to-image benchmarks.
The paper proposes a principled solution to the instability of end-to-end training in latent generative models. The core insight is identifying "prior-entropy imbalance" as the cause of latent collapse, where the generative objective's prior-fitting force overwhelms the weak KL regularization typically used in VAEs. The authors introduce an explicit entropy term to counteract this. Furthermore, they propose "GenFirst," a two-stage training schedule that prioritizes generation to shape the latent space before strengthening reconstruction. This addresses the asymmetric learning dynamics between the two objectives. The method is validated on both exact-likelihood autoregressive models (EAR) and flow-matching models (SiT/MMDiT), demonstrating generality.
The experimental results are strong. The paper reports a gFID of 0.97 on ImageNet-256 with SiT, which is a state-of-the-art result for diffusion models without using Fréchet Distance loss. It also achieves a GenEval score of 0.90 on text-to-image generation, outperforming larger models like FLUX.2-dev. The ablation studies are thorough, clearly isolating the effects of the entropy term and the GenFirst schedule. The comparison with REPA-E is particularly relevant, showing consistent improvements.
The paper provides detailed descriptions of the training schedules, loss weights, and architectural choices. However, specific hyperparameters for the "prior-only" phase and some implementation details of the GMM head are left to appendices or referenced works. The code availability is not explicitly stated in the provided text, which is a minor concern for immediate reproducibility, though the method is described clearly enough to be implemented.
The trade-off between generation and reconstruction is not fully eliminated; reconstruction fidelity (PSNR) still drops compared to standard VAEs. The autoregressive model (EAR) suffers from issues with Classifier-Free Guidance (CFG) scaling. The text-to-image experiments use a smaller dataset than industrial standards, so the scalability of the data efficiency claim is not fully tested at the largest scales.
This work provides a practical recipe for stable end-to-end latent learning, which could simplify the training pipeline for future generative models. By showing that latent spaces can be jointly optimized for generation and representation learning, it opens avenues for unified models that do not require separate pre-training stages for the tokenizer. The insights into entropy preservation are likely to be adopted in other variational frameworks. The paper identifies the root cause of latent collapse in end-to-end training (prior-entropy imbalance) and proposes a simple, effective strategy (GenFirst) to resolve the generation-reconstruction conflict, achieving state-of-the-art results on ImageNet and text-to-image benchmarks.
The integration of novel view synthesis (NVS) and open-vocabulary segmentation (OVS) has recently yielded powerful feed-forward 3D foundation models. However, their inherent reliance on static-scene assumptions leads to severe misalignment of spatial features in unconstrained dynamic environments. To bridge this critical gap, we propose SPAR, a novel joint semantic-geometric encoding architecture that explicitly isolates transient dynamic noise prior to latent space aggregation. Furthermore, we introduce a dynamic-region-aware end-to-end training paradigm that structurally couples motion estimation with multi-view visual and semantic learning. This unified approach enables the network to inherently resolve motion conflicts and distill multi-view consistent, temporally stable scene representations from dynamic inputs. Extensive experiments on the challenging D-RE10K benchmark demonstrate that SPAR achieves state-of-the-art performance. Our end-to-end approach achieves exceptional novel view synthesis quality, yielding a PSNR of 22.15 dB and 23.33 dB given only 3 and 4 input views respectively. Despite being trained in a self-supervised manner, our model achieves an mIoU of 88.5% for motion mask prediction. Furthermore, our analysis reveals a strong inter-task synergy between photometric scene reconstruction and semantic understanding, where semantic synthesis learning consistently enhances photometric fidelity in novel view rendering. Code will be available at https://github.com/dmucby/SPAR.
Primary: Institute of Automation, Chinese Academy of Sciences (CASIA)
All Institutions: ShanghaiTech University, Institute of Automation, Chinese Academy of Sciences (CASIA), The Chinese University of Hong Kong, Deepeleph Intelligent Technology
SPAR introduces a dynamic-robust photometric-semantic reconstruction framework that effectively isolates dynamic noise to enhance open-vocabulary 3D scene understanding in unconstrained environments. The paper presents a coherent solution to a significant problem in 3D vision, offering a unified training paradigm that improves both geometric and semantic consistency, though its impact is tempered by the specific benchmark usage and moderate photometric metrics.
The paper proposes SPAR, a joint semantic-geometric encoding architecture designed to handle dynamic scenes in open-vocabulary 3D scene understanding. The core methodological contribution is the explicit isolation of transient dynamic noise prior to latent space aggregation, which addresses the misalignment issues inherent in static-scene assumptions. Additionally, the authors introduce a dynamic-region-aware end-to-end training paradigm that couples motion estimation with multi-view visual and semantic learning. This unified approach allows the network to resolve motion conflicts and distill temporally stable representations. The methodology is logically sound and directly addresses a known limitation in feed-forward 3D foundation models.
Experiments are conducted on the D-RE10K benchmark. The reported results include a PSNR of 22.15 dB (3 views) and 23.33 dB (4 views) for novel view synthesis, and an mIoU of 88.5% for motion mask prediction. The paper claims state-of-the-art performance and highlights a synergy between photometric reconstruction and semantic understanding. However, the PSNR values are relatively modest for high-quality NVS, and the reliance on a single benchmark limits the generalizability of the claims. The self-supervised nature of the motion mask prediction is a strong point, but the absolute performance metrics need to be weighed against recent competitors in dynamic NVS.
The authors state that code will be available at the provided GitHub URL. The paper provides a clear description of the architecture and training paradigm. However, without access to the code or detailed hyperparameter settings in the text, full reproducibility is currently pending. The use of a specific benchmark (D-RE10K) aids in standardization.
The primary limitation is the reliance on the D-RE10K benchmark, which may not cover all dynamic scene complexities. The PSNR scores, while claimed to be SOTA, are not exceptionally high, suggesting potential room for improvement in photometric fidelity. The paper does not extensively discuss computational cost or inference speed, which are critical for real-time applications. Additionally, the "open-vocabulary" aspect is mentioned in the title but the depth of the semantic evaluation beyond mIoU is not fully detailed in the abstract.
This work contributes to the robustness of 3D foundation models in real-world, dynamic environments. By addressing the static-scene assumption, it enables more reliable applications in autonomous driving, robotics, and AR/VR where dynamic objects are prevalent. The synergy between semantic and geometric tasks offers insights into multi-task learning in 3D vision. SPAR introduces a dynamic-robust photometric-semantic reconstruction framework that effectively isolates dynamic noise to enhance open-vocabulary 3D scene understanding in unconstrained environments. The paper presents a coherent solution to a significant problem in 3D vision, offering a unified training paradigm that improves both geometric and semantic consistency, though its impact is tempered by the specific benchmark usage and moderate photometric metrics.
We present Alignment-Free Text-Audiobox (Text-AB), a unified framework for high-quality voice dubbing and full-duplex dialogue synthesis. Building on a Diffusion Transformer trained with a flow-matching objective, Text-AB departs from the Audiobox system along three dimensions. First, it operates in a latent diffusion framework using DAC-VAE features that encode 48 kHz waveforms into a 25 Hz latent sequence, giving over 10x higher compression than previous EnCodec representations while improving resynthesis quality. Second, Text-AB is alignment-free: it consumes raw text via an off-the-shelf text encoder and learns text-speech alignment through cross-attention, removing the need for forced alignment and explicit duration prediction. Third, we scale model and data substantially, pretraining a 3B-parameter model on 480k hours of monolingual speech, followed by supervised fine-tuning on three downstream tasks: cross-lingual voice dubbing, full-duplex dialogue synthesis, and emotional full-duplex dialogue synthesis. At inference, Text-AB supports one-shot generation for up to ~1 min of speech and arbitrarily long-form generation via a multi-diffusion scheme, plus a multi-stage reranking strategy that enhances quality based on automated metrics. On a real-world dubbing benchmark, Text-AB delivers a step-change improvement over the latest internal dubbing system, with large gains in prosody similarity, voice similarity, naturalness, and shareability. For full-duplex dialogue synthesis, it approaches human recordings on short-form conversations and substantially outperforms the latest internal model on long-form human-likeness and expressivity, while natively modeling turn-taking, back-channeling, and emotional dynamics. For emotional dialogue synthesis, emotion conditioning significantly improves emotion alignment and emotional interaction quality over the unconditioned baseline.
Primary: Meta AI
All Institutions: Meta AI
The paper presents a highly effective, scalable, and unified framework for advanced speech synthesis tasks, demonstrating state-of-the-art performance in voice dubbing and full-duplex dialogue generation through the combination of latent diffusion, alignment-free text conditioning, and large-scale training.
The paper proposes Text-AB, a unified framework for voice dubbing and full-duplex dialogue synthesis. The core methodological contributions are: (1) shifting from EnCodec to DAC-VAE latent features for higher compression and fidelity; (2) an "alignment-free" architecture that uses raw text embeddings via cross-attention instead of forced alignment and explicit duration prediction; and (3) a multi-stage training pipeline (pretraining on monologues, SFT for dubbing, SFT for dialogue) on a 3B parameter Diffusion Transformer. The use of flow-matching is standard in recent high-fidelity TTS (e.g., VoiceBox, Audiobox), and the latent diffusion approach is also established. The novelty lies primarily in the integration of these components into a single scalable system that handles both single-speaker dubbing and two-channel full-duplex dialogue without architectural changes other than input/output projection adjustments. The "alignment-free" aspect is a significant practical improvement, removing a brittle dependency on forced aligners.
The evaluation is extensive, covering three distinct tasks: cross-lingual voice dubbing, full-duplex dialogue synthesis, and emotional dialogue synthesis. The results show substantial improvements over internal baselines, particularly in human-likeness and prosody similarity. The use of both objective metrics (WER, SpkSim, Aesthetics) and subjective human evaluations (MOS, pairwise comparisons) provides a robust assessment. The ablation studies on model size, training data, and inference strategies (multi-diffusion, reranking) are thorough and demonstrate clear scaling laws and the importance of the proposed inference techniques. The comparison to ground-truth data for short-form dialogue is particularly compelling, showing near-human parity.
The paper provides detailed descriptions of the architecture, training data composition (hours, languages), hyperparameters (learning rates, batch sizes, GPU counts), and inference settings (ODE steps, reranking size). However, as is common with large-scale industrial papers, the specific datasets are internal and not publicly released. The code is not mentioned as being released. The reliance on internal benchmarks limits external reproducibility, but the methodological details are sufficient for a skilled practitioner to attempt a reproduction with similar data.
The primary limitation is the lack of public code and data, which hinders independent verification. The model is large (3B parameters) and requires significant computational resources for training and inference (256 A100s for training, multiple candidates for reranking). The evaluation is largely comparative against internal baselines, making it difficult to compare directly with other state-of-the-art open-source systems. The "alignment-free" approach relies on a powerful text encoder (mT5), which adds complexity and potential latency.
This work has significant implications for the development of full-duplex speech language models (FD-SLMs), which are a key component of next-generation voice assistants. By providing a high-quality, controllable method for generating synthetic full-duplex dialogue data, Text-AB can help address the data scarcity problem in training FD-SLMs. The unified framework for dubbing and dialogue synthesis also has practical applications in media production and accessibility. The alignment-free design simplifies the TTS pipeline, making it more robust to noisy or conversational speech. The paper presents a highly effective, scalable, and unified framework for advanced speech synthesis tasks, demonstrating state-of-the-art performance in voice dubbing and full-duplex dialogue generation through the combination of latent diffusion, alignment-free text conditioning, and large-scale training.
Accountability means a decision can be examined, justified, and contested. LLMs make this hard: fluent output may be ungrounded, incomplete, or unfaithful to the decision process. Achieving accountability requires verified rationales (how was the decision reached), assumptions (what was assumed rather than known), policy consistency (the same treatment for the same facts), and pivotal conditions (what would change the outcome). We introduce self-faithfulness as an automatic test of accountability: changing the pivotal conditions should change the decision. We examine accountable AI through clinical trial matching, a high-stakes task central to evidence-based medicine. Although LLM-based matchers match patients to trials reasonably accurately, they apply decision policies inconsistently and produce rationales that are unfaithful to their own decisions. We introduce VERDICT, an LLM-based agent that translates a decision task, its constraints, and its policy into Satisfiability Modulo Theories (SMT), then derives the decision with SMT and MaxSMT solvers -- so policies are applied consistently and decisions are accountable by construction. Across a SIGIR 2016-derived dataset and TREC 2021, VERDICT achieves the strongest decision accuracy among LLM-only and neurosymbolic baselines, applies policies with perfect consistency, and produces clinician-preferred rationales grounded in explicit assumptions and pivotal conditions, with improved counterfactual self-faithfulness.
Primary: Stanford University
All Institutions: Stanford University, Emory University
VERDICT introduces a neurosymbolic framework that translates LLM-based decision tasks into SMT formulas to ensure policy consistency and accountability, demonstrating superior accuracy and rationale fidelity in clinical trial matching compared to LLM-only and other neurosymbolic baselines.
The paper introduces VERDICT, a neurosymbolic agent that addresses the "accountability" gap in LLM-based decision-making. The core methodological contribution is the translation of natural language decision tasks, constraints, and policies into Satisfiability Modulo Theories (SMT) formulas. By using SMT and MaxSMT solvers to derive the final decision, the system ensures that the decision is logically consistent with the stated policy, unlike standard LLMs which may hallucinate or apply policies inconsistently. The introduction of "self-faithfulness" as a metric—testing whether changing pivotal conditions changes the decision—is a clever and rigorous way to evaluate the fidelity of the rationale to the underlying logic. The approach effectively bridges the gap between the flexibility of LLMs (for parsing and rationale generation) and the rigor of symbolic solvers (for decision consistency).
The experiments are conducted on two relevant datasets: a SIGIR 2016-derived dataset and TREC 2021, focusing on clinical trial matching. The evaluation metrics include decision accuracy, policy consistency, and clinician preference for rationales. The results claim that VERDICT achieves the strongest decision accuracy among LLM-only and neurosymbolic baselines and, crucially, achieves "perfect consistency" in policy application. The inclusion of human evaluation (clinician preferences) adds significant weight to the findings, as it validates the usability of the generated rationales in a high-stakes domain. The comparison against both pure LLM and other neurosymbolic baselines provides a comprehensive view of the method's superiority.
The paper provides a GitHub repository containing code, prompts, and evaluation artifacts, which significantly enhances reproducibility. The detailed appendix sections (including prompt search, parsing audits, and computational experiments) suggest a thorough implementation. However, the reliance on specific SMT solver configurations and the complexity of the translation pipeline from natural language to SMT may pose challenges for full reproduction without deep expertise in both NLP and formal methods.
The primary limitation is the dependency on the quality of the SMT translation. If the LLM fails to correctly translate the natural language policy into SMT, the downstream decision will be logically consistent but factually wrong. The paper likely discusses this in the limitations section, but the "perfect consistency" claim is only as good as the initial translation step. Additionally, the approach may not scale to highly complex, non-decomposable decision tasks where SMT encoding becomes intractable or overly complex. The evaluation is limited to clinical trial matching, so generalizability to other domains is not directly proven.
This work has significant implications for high-stakes AI applications where accountability and consistency are paramount, such as healthcare, finance, and legal tech. By providing a framework for "accountable by construction" AI, it offers a path toward regulatory compliance and trust in LLM-based systems. The concept of self-faithfulness could be adopted as a standard evaluation metric for explainable AI systems. The collaboration between Stanford and Emory suggests strong interdisciplinary impact, particularly in clinical AI. VERDICT introduces a neurosymbolic framework that translates LLM-based decision tasks into SMT formulas to ensure policy consistency and accountability, demonstrating superior accuracy and rationale fidelity in clinical trial matching compared to LLM-only and other neurosymbolic baselines.
Most prior works focused on conflicts between an LLM's internal parametric knowledge and externally provided context. In contrast, we investigate how LLMs handle conflicts that arise within contextual knowledge itself. We introduce a taxonomy of six types of contextual conflicts (factual, inferential, temporal, granularity, perspective, and ambiguity) and contribute a comprehensive dataset ContextConflict for this setting. The dataset contains 5,781 samples, covers both reasoning and summarization tasks, and includes both explicit contradictions and implicit conflicts that require multi-step reasoning. Experiments on nine LLMs show that current models still fall short in resolving contextual knowledge conflicts. We further provide mechanistic interpretability insights into how LLMs process such conflicts, revealing their latent awareness of conflicts and the representational geometry underlying conflict processing. In addition, our analysis uncovers a consistent model bias towards earlier evidence, and this positional preference serves as a key obstacle to effective conflict resolution. Motivated by these findings, we further propose a simple training-free, label-free steering method that steers activations to encourage a more comprehensive incorporation of evidences for better conflict resolution. On our dataset, the method consistently improves accuracy on reasoning tasks and generates higher-quality, more balanced summaries for summarization tasks.
Primary: Northwestern University
All Institutions: Northwestern University, New York University, University of Florida
The paper presents a comprehensive study on contextual knowledge conflicts in LLMs, contributing a new dataset, a mechanistic interpretability analysis, and a training-free steering method that effectively mitigates positional bias in evidence integration.
The paper introduces a well-structured taxonomy of six contextual conflict types (factual, inferential, temporal, granularity, perspective, ambiguity) and a corresponding dataset, ContextConflict. The methodology for dataset construction is rigorous, utilizing a mix of semi-synthetic generation (using GPT-5 for evidence perturbation while preserving gold labels from source datasets) and manual re-annotation to ensure quality. The interpretability analysis is a strong component, employing concept activation vectors (CAVs) to map conflict awareness across layers and spectral energy decomposition to analyze the geometric representation of conflicts. The proposed intervention, a training-free activation steering method, is logically derived from the finding of positional bias in the residual stream. By computing a steering direction that nudges activations toward a "neutral center" of evidence integration, the method directly addresses the identified mechanism of failure.
The experiments cover seven LLMs, including both closed-source (GPT-5, Claude, Gemini) and open-source models. The evaluation metrics are appropriate for the task: Accuracy for reasoning tasks and a novel Shapley-based Balance metric for summarization tasks. The validation of the Balance metric through human agreement studies and causal interventions adds significant rigor. The results demonstrate that current models struggle with these conflicts, particularly showing a strong bias toward earlier evidence. The proposed steering method yields consistent improvements in reasoning accuracy and summarization balance, validating the mechanistic insights. The ablation studies comparing the proposed steering direction with generic Context-Aware Steering (CAS) effectively isolate the contribution of the specific geometric correction.
The paper provides high reproducibility. The dataset is released on Hugging Face, and the code is available on GitHub. The prompts used for data generation are included in the appendix, and the specific models and hyperparameters for the steering method are detailed. The use of standard interpretability tools (CAVs, Shapley values) allows for independent verification of the analysis.
The primary limitation is the reliance on white-box access for the steering method, restricting its application to open-weight models. The paper acknowledges that scaling to larger models (70B+) is left for future work due to compute constraints. Additionally, the dataset, while comprehensive, still relies on semi-synthetic construction for some conflict types, which may not fully capture the noise of real-world retrieval-augmented generation scenarios. The Shapley-based metric assumes equal contribution of evidence pieces in expectation, which may not hold in all real-world settings.
This work contributes to the understanding of LLM reliability in multi-document settings, which is critical for applications in legal, medical, and news aggregation. The identification of positional bias in conflict resolution provides a clear target for future alignment techniques. The dataset serves as a valuable benchmark for evaluating model robustness against conflicting information, moving beyond simple factual contradiction to more nuanced types of conflict. The paper presents a comprehensive study on contextual knowledge conflicts in LLMs, contributing a new dataset, a mechanistic interpretability analysis, and a training-free steering method that effectively mitigates positional bias in evidence integration.
The growing rate at which LLM agents interact with one another raises key questions about language evolution in multi-LLM-agent settings, with implications for safety and monitorability as well as for linguistic accounts of LLMs. To address these questions, we introduce GlossoGen, a novel platform for studying multi-agent language evolution in complex scenarios. Within GlossoGen, we build the SaveVeyru scenario, which requires agents with partial information to communicate under pressure. We find that language evolution does occur between LLM agents, that the resulting languages are compositional and morphologically productive, and that they deviate from the LLMs' English prior in ways that render them incomprehensible to humans. Moreover, we identify several qualities essential to this evolution: pressure towards efficiency; the strength of the models backing the agents; and access to a "postmortem" stage in which agents can agree on linguistic conventions. Importantly, we observe that different conditions govern the transmission of language to new agents. Specifically, we find that agents learn new languages from usage alone, take an active role in this learning, and that while stronger models are required for novel language emergence, weaker models can learn an existing language once it has emerged. Taken together, our results indicate that current LLMs have the potential for cumulative cultural evolution -- previously attested only in humans -- with mixed populations of agents developing capacities that go beyond their lowest common denominator.
Primary: University of Cambridge
All Institutions: University of Cambridge, Anthropic, OpenAI
The paper presents a rigorous and novel framework for studying emergent language in LLM agents, demonstrating that under specific pressures, agents develop compositional, non-English languages that can be transmitted across model generations, highlighting both the capabilities and safety risks of multi-agent LLM systems.
The paper introduces GlossoGen, a flexible platform for simulating multi-agent LLM interactions with a focus on language evolution. The core methodological contribution is the "SaveVeyru" scenario, which forces information asymmetry and communication efficiency through a character budget. By using a fictional alien entity, the authors successfully prevent LLMs from relying on parametric medical knowledge, ensuring that communication is necessary for task success. The methodology includes rigorous controls for "postmortem" deliberation phases and budget pressures. The analysis of emergent languages employs linguistic tools, such as perplexity under an English LM and morphosyntactic induction via LLM judges, to characterize the structure of the evolved codes.
The experiments are extensive, testing both proprietary (GPT-5.4, Claude Opus 4.7, Sonnet 4.6) and open-weight (Llama, Qwen) models. Key findings include: (1) Language emergence requires both budget pressure and a postmortem phase; (2) Emergent languages are compositional and morphologically productive, not just memorized codebooks; (3) Stronger models are required to *create* new languages, but weaker models can *learn* them from usage, suggesting a path to cumulative cultural evolution. The statistical analysis (mixed-effects models) supports the significance of these findings. The observation that agents ask metalinguistic questions to clarify atomic terms but not compositional ones is a particularly insightful result regarding LLM internal representations.
The authors provide a GitHub repository with the platform and scenario definitions. The use of standard LLM APIs ensures that the core experiments are reproducible, though exact replication of proprietary model behaviors may vary with model updates. The detailed description of the scoring mechanism (LLM judge verified against gold data) enhances trust in the results.
The study relies on a single, albeit complex, scenario (SaveVeyru). It is unclear if these findings generalize to other types of multi-agent tasks (e.g., competitive vs. cooperative, or different domain structures). The "postmortem" phase is a strong confounder; in real-world dynamic environments, such explicit deliberation time may not be available. Additionally, the claim of "cumulative cultural evolution" is based on short-term transmission within a simulation, not long-term open-ended evolution.
This paper has significant implications for AI safety, demonstrating that LLM agents can develop unmonitorable communication protocols even in cooperative settings without adversarial pressure. It challenges the assumption that LLMs will always communicate in human-readable languages. For linguistics and cognitive science, it provides a new testbed for studying language emergence and transmission in artificial agents, potentially offering insights into the cognitive prerequisites for language evolution. The paper presents a rigorous and novel framework for studying emergent language in LLM agents, demonstrating that under specific pressures, agents develop compositional, non-English languages that can be transmitted across model generations, highlighting both the capabilities and safety risks of multi-agent LLM systems.
Modern systems in artificial intelligence (AI) somehow excel in domains for which they seem poorly suited. Intelligence has traditionally been modeled as operating over structured combinations of symbols, such as logical formulas. However, the strongest modern AI systems are based on neural networks, which instead represent information in continuous vectors. Vectors seem inadequate for capturing the structure of language, logic, and other cognitive domains, yet neural networks achieve impressive performance in these areas. How do they do it? In this work, we propose a potential answer: Despite appearances, perhaps the internal representations of neural networks implicitly realize symbolic structure. In support of this hypothesis, we show that the vector representations of a variety of neural networks can be closely approximated with symbolic structures: we can replace the network's entire representation-generating process with a closed-form equation instantiating a symbolic structure, and the network's behavior remains largely unchanged. This finding holds for both small-scale neural networks trained to manipulate lists as well as large language models (LLMs) operating in four domains that are central in symbolic traditions: arithmetic, logic, computer code, and language. Further, our symbolic approximation allows us to modify an LLM's behavior in targeted ways via precise interventions on its internal representations, showing that the LLM's behavior is reliant on the symbolic structures we have identified. This work provides a potential way to reconcile longstanding symbolic conceptions of intelligence with the vector-based nature of modern AI.
Primary: Johns Hopkins University
All Institutions: Johns Hopkins University, Yale University, New York University, Microsoft Research, Princeton University
The paper demonstrates that neural networks, including large language models, implicitly use Tensor Product Representations to encode symbolic structure. By introducing the DISCOVER method, the authors provide a rigorous framework for extracting and validating these structures, showing that they are not only present but causally responsible for model behavior in arithmetic, logic, coding, and language tasks.
The paper proposes DISCOVER (DISsecting COmpositionality in VEctor Representations), a method to test the hypothesis that neural networks implicitly realize symbolic structure via Tensor Product Representations (TPRs). The core technique involves training a parametric TPR model to approximate the internal vector representations of a target neural network (black-box). The success of this approximation is evaluated by feeding the TPR-generated vectors into the target network's decoder (or a trained period-unpacking model for LLMs) and measuring task accuracy. The methodology is rigorous, leveraging established cognitive science formalisms (TPRs) to provide a structured, interpretable alternative to the "Linear Representation Hypothesis." The use of "linearly-transformed TPRs" adds necessary flexibility to fit real-world neural network geometry.
The experiments are extensive, covering small-scale synthetic tasks (copying, reversing, interleaving lists) across four architectures (MLP, GRU, Transformer, Bottleneck Transformer) and seven large language models (Gemma-3, GPT-2-XL, GPT-OSS, Pythia, Qwen3, OLMo-2, Llama-3.1). The paper demonstrates that bidirectional role schemes consistently provide high-fidelity approximations of network representations, whereas bag-of-words (structureless) schemes fail. A particularly strong result is the intervention study, where editing the TPR representation (e.g., swapping subject/object roles) successfully alters the LLM's behavior in the expected way, confirming the causal relevance of the identified structure. The finding that DISCOVER approximations often outperform the actual noisy LLM representations in downstream decoding tasks is a significant empirical insight.
The authors provide a partial codebase on GitHub. The paper details the architecture specifications, training hyperparameters, and dataset generation procedures in the appendices. The use of open-weights LLMs from Hugging Face ensures that the models analyzed are accessible. However, the "partial" nature of the code and the complexity of the TPR fitting procedure may pose some barriers to exact reproduction without the full implementation details.
The analysis primarily focuses on the representation of the final token (period) in LLMs, which may not capture the full dynamics of intermediate layers or token-level representations for all tasks. The role schemes are hypothesized by the experimenter (supervised DISCOVER), which limits the method's ability to discover unknown structures autonomously. The paper acknowledges that TPRs are a specific formalism and does not rule out other potential symbolic structures, though it argues for the generality of TPRs.
This work bridges the gap between symbolic AI and connectionist AI, offering a concrete mechanism for how neural networks handle structured information. It provides a new tool for mechanistic interpretability, allowing researchers to extract interpretable, closed-form equations for network representations. This could facilitate better alignment, debugging, and control of LLMs by enabling precise interventions on symbolic features. It challenges the notion that neural networks are purely "bag-of-words" systems and reinforces the importance of structure in intelligent behavior. The paper demonstrates that neural networks, including large language models, implicitly use Tensor Product Representations to encode symbolic structure. By introducing the DISCOVER method, the authors provide a rigorous framework for extracting and validating these structures, showing that they are not only present but causally responsible for model behavior in arithmetic, logic, coding, and language tasks.
Speech Language Models (SLMs) are increasingly deployed in multi-speaker environments, yet their ability to attribute speech to the correct speaker and reason over speaker identities remains unclear. Hence, we introduce HEAR, a conceptually hierarchical benchmark diagnosing the foundational capabilities of speaker-attributed reasoning, comprising 2.4K human-verified samples from 887 diverse multi-party audio clips. Evaluating 20 leading SLMs on HEAR reveals they struggle with these foundational tasks, often relying on semantic priors rather than actual vocal cues. To address this, we present A2R, a 30B model optimized on Counterfactual Audio with Speaker-level Hard negatives (CASH), a dataset designed to guide the model to prioritize acoustic vocal cues over linguistic signals. A2R achieves strong performance on HEAR and exhibits zero-shot generalization to diverse multi-speaker downstream tasks, demonstrating that learned speaker attribution unlocks the model's latent capacity for speaker-aware reasoning. All resources are available at https://attributetoreason.github.io/AttributeToReason/
Primary: Seoul National University
All Institutions: Seoul National University, Samsung Electronics Co., Ltd., NVIDIA
The paper introduces the HEAR benchmark and the A2R model, demonstrating that speaker-attributed reasoning in Speech Language Models can be enhanced by training on counterfactual audio that decouples linguistic content from speaker identity. This work provides a critical diagnostic tool for the field and a viable path toward more robust, acoustically-grounded multimodal reasoning, addressing a fundamental limitation in current SLMs.
The paper introduces a two-pronged approach: a diagnostic benchmark (HEAR) and a specialized model (A2R). The benchmark is hierarchical, targeting the specific failure mode of Speech Language Models (SLMs) in multi-speaker environments: the inability to distinguish between semantic priors and actual acoustic speaker cues. The core methodological innovation lies in the training data for A2R, specifically the "Counterfactual Audio with Speaker-level Hard negatives" (CASH) dataset. By constructing counterfactual audio where the linguistic content is decoupled from the speaker identity (likely via voice cloning or TTS), the authors force the model to rely on paralinguistic and acoustic features rather than textual context. This is a sophisticated data-centric approach to solving a reasoning problem. The 30B parameter scale of A2R suggests a significant computational investment, likely fine-tuning a large multimodal foundation model. The methodology is sound, directly addressing the identified gap in current SLM capabilities.
The evaluation is extensive, testing 20 leading SLMs on the new HEAR benchmark. The finding that these models rely on semantic priors is a significant empirical contribution, as it quantifies a known but previously under-diagnosed weakness. The performance of A2R is reported to be strong, with zero-shot generalization to downstream tasks. However, the provided text is a summary/abstract-like structure rather than the full body, so specific numerical comparisons (e.g., accuracy percentages, ablation studies on the CASH dataset components) are not visible. The claim of "zero-shot generalization" is a strong indicator of robustness, but without seeing the specific downstream tasks and baseline comparisons in detail, the magnitude of improvement is inferred from the abstract's confidence. The use of human-verified samples (2.4K) adds credibility to the benchmark's quality.
The authors state that all resources are available at the provided URL. The ethics statement indicates that synthetic waveform data is restricted to a gated repository due to voice cloning risks, which limits full reproducibility of the training data for the general public. However, code, evaluation protocols, and non-identifying annotations are public. This is a reasonable balance for this type of research, though it does hinder independent verification of the CASH dataset construction by external parties without access.
The primary limitation is the latency introduced by explicit reasoning (transcript generation) in A2R, which hinders real-time application. The authors acknowledge this and suggest implicit reasoning as future work. Additionally, the reliance on voice cloning for the CASH dataset raises ethical and legal concerns regarding consent, which the authors address with a strict Data Use Agreement, but this restricts the open availability of the core training data. The benchmark size (2.4K samples) is moderate; while high-quality, it may not capture the full diversity of real-world multi-speaker interactions.
This work has significant implications for the development of trustworthy multi-party conversational AI. By providing a benchmark that exposes the "semantic prior" bias in SLMs, it guides the field toward more robust acoustic grounding. The A2R model demonstrates that speaker attribution can be explicitly learned, which is crucial for applications in meeting assistants, collaborative robots, and accessible communication tools. The ethical framework provided for voice cloning research is also a valuable contribution to the community's standards. The paper introduces the HEAR benchmark and the A2R model, demonstrating that speaker-attributed reasoning in Speech Language Models can be enhanced by training on counterfactual audio that decouples linguistic content from speaker identity. This work provides a critical diagnostic tool for the field and a viable path toward more robust, acoustically-grounded multimodal reasoning, addressing a fundamental limitation in current SLMs.
Scientific progress depends not only on finding solutions, but on learning the rules that explain why they work and using that understanding to design better experiments. We introduce science sandboxes, a framework for studying this capability in AI agents through repeated cycles of experimentation, feedback, and hypothesis revision. Science sandboxes invite an agent to query the natural world in different ways, ranging from "wet" physical experiments, to "damp" predictive models trained on empirical data, to "dry" invented rules. By establishing a common experimental loop and a protocol for evaluating agents within it, science sandboxes allow assessment of both quantitative performance on specific metrics and qualitative scientific reasoning, across a spectrum of empirical verifiability. Here, we instantiate this framework in two biological settings, models of regulatory genomics and protein fitness prediction, and examine the capabilities of frontier agents. Across these settings, we could see when agents successfully optimized a quantitative metric without understanding the rules underlying the system. In particular, their scientific reasoning deteriorated when they encountered systems whose rules fell outside familiar biological priors. By highlighting such failure modes, science sandboxes make the frontier of scientific capability measurable and provide a controlled setting in which to study and ultimately expand it.
Primary: The Broad Institute of MIT and Harvard
All Institutions: The Broad Institute of MIT and Harvard, Harvard Medical School, The Jackson Laboratory, Sutter Hill Ventures, David H. Koch Institute for Integrative Cancer Research, Howard Hughes Medical Institute, The Wyss Institute for Biologically Inspired Engineering at Harvard University, Harvard-MIT Program in Health Sciences and Technology, Yale School of Medicine, Department of Systems Biology, Department of Biology, Department of Immunology and Infectious Diseases, Department of Organismic and Evolutionary Biology
The paper introduces "science sandboxes," a novel framework for evaluating AI agents' scientific reasoning by distinguishing between quantitative optimization and qualitative rule discovery. Through rigorous experiments in regulatory genomics and protein folding, it demonstrates that while frontier agents can match human performance on familiar tasks, they struggle to infer underlying rules when faced with systems outside their pretraining priors, providing a critical benchmark for the development of truly scientific AI agents.
The paper introduces "science sandboxes," a framework for evaluating AI agents' scientific reasoning capabilities by distinguishing between quantitative optimization and qualitative rule discovery. The methodology is robust, defining a clear experimental loop (specimens, assays, oracle) and categorizing oracles into "wet" (physical), "damp" (computational models), and "dry" (invented rules). The implementation of two specific sandboxes, MPRAbox (regulatory genomics) and CodonBox (protein folding/genetic code inference), is well-designed. MPRAbox uses a damp oracle based on the Malinois model to simulate MPRA experiments, while CodonBox uses a dry oracle with hidden, non-biological rules to test generalization beyond prior knowledge. The evaluation protocol includes both single-round and multi-round (long-horizon) tests, allowing for the assessment of hypothesis revision over time.
The experiments are extensive and insightful. The authors evaluate frontier agents (Claude Opus 4.7, GPT-5.5, Gemini 3.5 Flash) against human-designed baselines. Key findings include: (1) Agents can match or exceed human performance on standard biological tasks (MPRAbox) by leveraging pretraining priors. (2) Performance drops significantly when agents are forced to discover rules outside their biological priors (dry oracles in MPRAbox and CodonBox). (3) Multi-round experiments reveal that agents often engage in brute-force search rather than structured hypothesis testing, and they struggle to infer complex combinatorial rules (e.g., interactions between codon positions) when the search space grows. The qualitative analysis of agent "lab notebooks" provides strong evidence for the distinction between score optimization and true understanding.
The paper provides detailed descriptions of the sandbox mechanics, oracle rules (Table 3), and experimental protocols. However, as an arXiv preprint, the code and specific agent configurations are not yet publicly available. The reliance on specific frontier models (which change rapidly) limits long-term reproducibility, but the framework itself is clearly defined enough for others to implement similar sandboxes.
The evaluation is limited to two biological domains (genomics and protein folding), which may not generalize to all scientific fields. The "dry" oracles, while useful for testing generalization, are artificial and may not reflect the complexity of real-world scientific discovery. The paper focuses on a small number of frontier agents, and the results may not apply to smaller or specialized models. Additionally, the qualitative evaluation of lab notebooks is currently manual, though the authors suggest automation is possible.
This paper has high potential impact on the AI for Science community. It provides a rigorous framework for measuring "scientific capability" beyond mere benchmark scores, which is a critical gap in current AI evaluation. The findings highlight the limitations of current LLMs in genuine scientific induction, guiding future research in agent design and training. The concept of "science sandboxes" is likely to be adopted as a standard evaluation method for scientific AI agents. The paper introduces "science sandboxes," a novel framework for evaluating AI agents' scientific reasoning by distinguishing between quantitative optimization and qualitative rule discovery. Through rigorous experiments in regulatory genomics and protein folding, it demonstrates that while frontier agents can match human performance on familiar tasks, they struggle to infer underlying rules when faced with systems outside their pretraining priors, providing a critical benchmark for the development of truly scientific AI agents.
Generalizable embodied manipulation remains difficult to achieve through pretraining alone, due to unseen physical conditions in the real world. We argue that robots need to learn from their own physical interactions on the fly during real-world deployment and use this knowledge to inform subsequent actions. We present Zeva, the first framework that enables in-context learning from a robot's own physical interaction experience while keeping the policy model frozen. Zeva employs a Causal Interaction Extractor to encode an executed action and its induced state change into a causal interaction signal, which is stored in a dual-timescale causal memory. For subsequent actions, relevant causal interaction signals are retrieved from memory and injected into the frozen policy model as context. Experiments in simulation and real-world manipulation demonstrate that Zeva achieves the best performance among the compared frontier VLAs and WAMs and, more importantly, enables self-evolution during deployment without gradient updates. Its success rate continues to improve as the robot accumulates interaction experience. Furthermore, the acquired interaction experience can generalize across tasks.
Primary: Institute for AI Industry Research (AIR), Tsinghua University
All Institutions: Institute for AI Industry Research (AIR), Tsinghua University, Z-Trans AI
Zeva introduces a novel in-context causal learning framework that enables frozen robot policies to self-improve through accumulated interaction memory. The paper demonstrates that by extracting and retrieving causal interaction signals from past attempts, robots can significantly enhance their manipulation success rates in real-world settings without gradient updates, offering a scalable solution to the distribution shift problem in embodied AI.
The paper proposes Zeva, a framework for In-Context Causal Learning (ICCL) in embodied manipulation. The core innovation is the decoupling of policy adaptation from parameter updates. Instead of fine-tuning the Vision-Language-Action (VLA) model, Zeva extracts "causal interaction signals" from the robot's own execution history using a Causal Transition Encoder (CTE). These signals are stored in a dual-timescale memory: a Brief Interaction Trace (BIT) for immediate context and a Persistent Interaction Memory (PIM) for cross-attempt learning. The retrieved causal context is injected into the frozen policy's attention layers as a prompt. This approach allows the robot to learn from failures and successes in real-time without the computational cost or risk of catastrophic forgetting associated with gradient-based updates. The method is conceptually strong, leveraging the "test-time scaling" paradigm by treating interaction history as a form of in-context learning data.
The experiments are conducted on both simulation (RoboCasa365-Atomic5) and a real-world chemical lab benchmark (ChemLab-Evo). The real-world setup is particularly impressive, involving complex multi-stage tasks like titration and extraction. Zeva outperforms strong baselines like OpenVLA, Fast-WAM, and Cosmos3-Nano. The ablation studies convincingly demonstrate the necessity of both memory streams (BIT and PIM). The "Post-Deployment In-Context Scaling" experiments show monotonic improvement in success rates as the robot accumulates experience, validating the core hypothesis. The cross-task generalization analysis, where causal signals from one task successfully inform another, provides strong evidence that the CTE is learning functional physical dynamics rather than just memorizing task-specific trajectories.
The paper provides a clear algorithmic description and detailed architectural diagrams. However, specific hyperparameters for the CTE training, the exact structure of the "Cosmos3" backbone (which appears to be a proprietary or internal model given the citation style), and the implementation details of the memory retrieval mechanism are not fully specified in the text provided. The reliance on a specific, potentially non-public foundation model (Cosmos3) may limit immediate reproducibility for the broader community, although the framework itself is modular.
The primary limitation is the reliance on a specific foundation model (Cosmos3) which may not be publicly available, hindering independent verification. The method assumes that the CTE can be pre-trained or fine-tuned to extract meaningful causal signals; the paper does not deeply explore the cost or data requirements for training this encoder. Additionally, the "causal" label is somewhat aspirational; the method learns correlations between actions and state changes, but true causal inference (counterfactuals) is not explicitly demonstrated. The evaluation is limited to a single robot platform (ARX) and a specific set of lab tasks, so generalizability to other embodiments or highly dynamic environments remains to be seen.
This work addresses a critical bottleneck in embodied AI: the gap between pre-training data and real-world deployment conditions. By enabling self-evolution through in-context learning, Zeva offers a path toward robots that can adapt to new environments without retraining. This has significant implications for industrial automation and service robotics, where retraining models for every new task or environment is impractical. The concept of "causal memory" could be extended to other domains requiring long-horizon planning and adaptation. Zeva introduces a novel in-context causal learning framework that enables frozen robot policies to self-improve through accumulated interaction memory. The paper demonstrates that by extracting and retrieving causal interaction signals from past attempts, robots can significantly enhance their manipulation success rates in real-world settings without gradient updates, offering a scalable solution to the distribution shift problem in embodied AI.
Traversing sparse 3D structures requires humanoid robots to perceive thin, overhanging geometry while executing agile, accurate whole-body motions. We study this problem through monkey-bar traversal, where the robot must jump to the structure, traverse it through sparse bar interactions, and land safely. For this task, we present a reinforcement-learning-based perceptive control system that operates directly on observations from a head-mounted solid-state lidar. To extract task-relevant geometry from the sparse returns, the policy consumes the raw lidar scan through an attention-based encoder with recurrent memory. This policy is obtained by a phase-scheduled teacher- student pipeline that combines privileged experts for jumping up, brachiating, and jumping down. For transfer to hardware, we model lidar noise, battery-voltage sag, and actuator thermal limits, and equip the humanoid with passive hook end-effectors for robust bar interaction. On hardware, the resulting policy completes the full jump-up->brachiation->jump-down sequence in 14 of 15 trials across three bar configurations and reaches brachiation speeds up to 0.5 m/s. Beyond brachiation, the same perception backbone supports a separately trained policy that ducks beneath thin overhead obstacles with 2 cm cross-sections.
Primary: ETH Zurich
All Institutions: ETH Zurich, ETH AI Center
The paper presents a novel perceptive control system enabling humanoids to traverse sparse 3D structures via brachiation using raw lidar. It contributes a rigorous sim-to-real framework that models battery sag, thermal limits, and sensor-specific noise, achieving high success rates on hardware and demonstrating the viability of map-free perception for agile whole-body control.
The paper proposes a reinforcement learning framework for humanoid brachiation (monkey bar traversal) that operates directly on raw solid-state lidar data, bypassing intermediate map representations like elevation maps or voxel grids. The core methodological contribution is a phase-scheduled teacher-student distillation pipeline. Three privileged teachers (for jumping up, brachiating, and jumping down) are distilled into a single student policy that uses an attention-based encoder (adapted from AME-2) and recurrent memory to process sparse lidar returns. A key technical strength is the rigorous sim-to-real transfer modeling, which explicitly accounts for battery voltage sag, actuator thermal limits, and specific lidar noise artifacts (ray divergence/edge bleed). The use of passive hook end-effectors is a clever hardware simplification that reduces control complexity.
The experiments are robust and well-structured. The authors provide ablations on the encoder architecture and auxiliary losses, demonstrating the necessity of the attention-based approach for sparse geometry. Sim-to-sim validation in MuJoCo with a high-fidelity lidar model precedes hardware testing. On hardware, the robot achieves a 93% success rate (14/15 trials) on the full jump-up/brachiation/jump-down sequence across three different bar configurations. The demonstration of a secondary task (ducking under thin obstacles) using the same perception backbone further validates the generalizability of the perceptual encoder.
The paper provides detailed descriptions of the reward functions, curriculum, and sensor noise models. Specific parameters for the battery and thermal models are discussed, and the lidar noise model is described with sufficient detail (beam divergence, edge dropout probabilities) to be replicated. The use of standard tools like IsaacLab and RSL-RL aids reproducibility, though the specific hardware modifications (passive hooks) and sensor mounting require physical access to the PM-01 platform.
The system is currently limited to a small set of separately trained policies for specific tasks (brachiation, ducking). The paper acknowledges that robustness to substantially more diverse geometry remains to be demonstrated. The reliance on a specific solid-state lidar (RoboSense E1R) and the specific hardware platform (PM-01) limits immediate generalizability to other humanoid platforms without re-tuning the sim-to-real models.
This work pushes the boundary of what is possible with legged robots in unstructured, sparse environments. By demonstrating that raw lidar can support explosive, contact-accurate motions, it challenges the prevailing paradigm of using dense intermediate maps for locomotion. The techniques for modeling actuator thermal limits and battery sag are broadly applicable to any high-power robotic system operating near its physical limits. The paper presents a novel perceptive control system enabling humanoids to traverse sparse 3D structures via brachiation using raw lidar. It contributes a rigorous sim-to-real framework that models battery sag, thermal limits, and sensor-specific noise, achieving high success rates on hardware and demonstrating the viability of map-free perception for agile whole-body control.
We present $\mathcal{N}_0$-Foundation, a paradigm for tactile-enabled embodied manipulation, which integrates tactile sensing hardware, large-scale multimodal data, tactile representation learning, and standardized evaluation. First, we engineer the infrastructure for scalable data collection, including a vision-based tactile sensor, a tactile Universal Manipulation Interface (UMI), and a synchronized visuo-tactile data collection system supporting both robot embodiments and UMI-based demonstrations. Leveraging this infrastructure, we construct NeoData, which contains more than 30000 hours of synchronized visual and tactile demonstrations, spanning six embodiments, 450 tasks, and billions of paired RGB and tactile frames collected through a mixture of real-robot teleoperation and UMI-based demonstrations. To facilitate open research, we further release OpenNeoData, a 5000-hour open-source subset of NeoData. The dataset addresses a central limitation of existing manipulation corpora, critical for deformable-object manipulation, precise assembly, delicate force control, and sustained surface interaction. Capitalizing on the large-scale, heterogeneous tactile measurements, we propose NeoForce, a visuo-tactile representation model that learn transferable tactile representations across different sensor designs. To enable systematic evaluation of tactile embodied models built upon our infrastructure, datasets and tactile representations, we further propose a comprehensive benchmark, which combines the real-world NeoReal suite and the simulated NeoSim suite for standardized evaluation. Experiments across both suites show that policies benefit from the physical contact state rather than from the device-specific appearance of the tactile signal. We release the dataset, the representation, and the benchmark, aiming at supporting future work on tactile-enabled embodied manipulation.
Primary: Fudan University
All Institutions: Fudan University, NeoteAI
The paper introduces a large-scale tactile dataset, a transferable visuo-tactile representation model, and a comprehensive benchmark, establishing a foundational infrastructure for tactile-enabled embodied manipulation. By addressing the critical data scarcity in tactile sensing and providing a standardized evaluation framework, the work enables significant progress in dexterous robotics and force control, offering a robust baseline for future research in multimodal manipulation.
The paper proposes a comprehensive paradigm for tactile-enabled embodied manipulation, comprising three main technical pillars: (1) Hardware Infrastructure: The development of a vision-based tactile sensor and a "tactile UMI" (Universal Manipulation Interface) that allows for synchronized visuo-tactile data collection from both robot embodiments and human demonstrations. This addresses the scarcity of high-quality tactile data by enabling scalable collection via low-cost, portable interfaces. (2) Data & Representation: The construction of NeoData (30,000+ hours) and the release of OpenNeoData (5,000 hours). The core algorithmic contribution is NeoForce, a visuo-tactile representation model designed to learn transferable tactile features across heterogeneous sensor designs. This is crucial because tactile signals are often device-specific; learning a canonical representation allows policies to generalize across different hardware. (3) Benchmarking: The introduction of NeoReal (real-world) and NeoSim (simulated) suites for standardized evaluation. The methodology is sound, leveraging the scale of the data to train robust representations, and the focus on transferability is a significant methodological advance over previous work that often treated tactile data as modality-specific noise.
The experimental section evaluates policies trained on the proposed representations across both real-world and simulated environments. Key findings indicate that policies benefit from the physical contact state encoded in the tactile representation rather than the device-specific appearance of the raw signal, validating the transferability claim. The scale of the dataset (billions of paired frames) is a major strength, providing a resource that was previously unavailable. The evaluation covers a wide range of tasks (450 tasks, 6 embodiments), demonstrating the generality of the approach. However, the specific quantitative gains over strong baselines (such as vision-only policies or non-transferable tactile policies) need to be scrutinized for statistical significance and practical relevance. The comparison with existing tactile datasets (like T-DEX or similar) is implied but should be explicit to highlight the unique value of the scale and modality pairing.
The paper explicitly commits to releasing OpenNeoData (5,000 hours), the NeoForce representation model, and the benchmark suites. This is a high level of reproducibility, especially given the difficulty of collecting tactile data. The hardware design details for the tactile UMI and sensor are provided, allowing other labs to replicate the data collection infrastructure. The code for the representation learning and policy training is expected to be released, which is standard for such large-scale contributions. The only barrier to full reproducibility is the cost and effort of building the custom hardware, but the open-source data mitigates this for algorithmic research.
The primary limitation is the hardware dependency; while the data is open, the specific tactile sensors used may not be easily accessible to all researchers, potentially limiting the immediate adoption of the full pipeline. The "tactile UMI" is a novel device, and its calibration and maintenance requirements are not fully detailed in the abstract. Additionally, the computational cost of processing billions of tactile frames and training large-scale multimodal models is significant. The paper may also face challenges in generalizing to tactile sensors with fundamentally different sensing principles (e.g., capacitive vs. vision-based), as the current focus is on vision-based tactile sensing.
This work has the potential to significantly accelerate research in dexterous manipulation and force-sensitive robotics. By providing a large-scale, open dataset and a transferable representation, it lowers the barrier to entry for developing tactile-aware policies. It bridges the gap between vision-centric manipulation and the need for physical interaction feedback. The standardization of benchmarks (NeoReal/NeoSim) will facilitate fair comparison of future tactile manipulation algorithms. This could lead to more robust and versatile robots capable of handling deformable objects and performing precise assembly tasks, which are currently difficult for vision-only systems. The paper introduces a large-scale tactile dataset, a transferable visuo-tactile representation model, and a comprehensive benchmark, establishing a foundational infrastructure for tactile-enabled embodied manipulation. By addressing the critical data scarcity in tactile sensing and providing a standardized evaluation framework, the work enables significant progress in dexterous robotics and force control, offering a robust baseline for future research in multimodal manipulation.
Conventional flow-based vision-language-action (VLA) models support expressive continuous action generation but rely on multi-step refinement to produce each action chunk, increasing latency in online robot control. To address this issue, we introduce DriftingVLA, a native one-step VLA that generates a complete action chunk with a single action-expert forward pass. Rather than learning a flow field that requires iterative integration at inference, DriftingVLA uses a distribution-drifting objective to learn a direct noise-to-action-chunk mapping for one-step deployment. Since robot action dimensions carry distinct control semantics and distributional characteristics, we further introduce Per-Dimension Temporal Drifting (PDTD). PDTD treats the complete temporal trajectory of each action dimension as a separate drifting unit, enabling finer-grained modeling and shaping of dimension-specific action distributions. This per-dimension decomposition applies only to the training objective; the shared VLA model still generates the complete action chunk jointly, thereby preserving cross-dimensional dependencies. DriftingVLA achieves 98.32% success on LIBERO, 81.09% on RoboTwin 2.0, and 77.67% across six real-world single- and dual-arm tasks, outperforming the evaluated multi-step flow policy and one-step VLA baselines. Native one-step deployment also delivers a 3.36-fold speedup in action-chunk generation, eliminating iterative refinement without sacrificing control performance.
Primary: University of Science and Technology of China
All Institutions: University of Science and Technology of China, LYNSENSE
DriftingVLA introduces a native one-step VLA generation method using Per-Dimension Temporal Drifting to achieve significant latency reduction without sacrificing performance. The paper presents a solid engineering and algorithmic contribution to the VLA field, offering a practical solution to the inference latency problem inherent in flow-based models, with strong empirical results on both simulation and real-world robotic tasks.
The paper proposes DriftingVLA, a Vision-Language-Action (VLA) model that replaces the iterative refinement of flow matching with a native one-step generation approach. The core innovation is the "distribution-drifting" objective, which trains a direct mapping from noise to action chunks, bypassing the need for multi-step numerical integration at inference. A key technical contribution is Per-Dimension Temporal Drifting (PDTD), which decomposes the training objective by treating the temporal trajectory of each action dimension as a separate drifting unit. This allows for finer-grained modeling of distinct control semantics (e.g., position vs. orientation) while maintaining a shared model architecture that generates the full action chunk jointly, thus preserving cross-dimensional dependencies. The method effectively addresses the latency bottleneck in online robot control by reducing the action-expert forward passes to one per chunk.
The experimental evaluation is robust, covering simulation benchmarks (LIBERO, RoboTwin 2.0) and real-world tasks (six single- and dual-arm tasks). The reported success rates are high: 98.32% on LIBERO, 81.09% on RoboTwin 2.0, and 77.67% on real-world tasks. The paper claims outperformance over both multi-step flow policies and other one-step VLA baselines. A significant practical advantage is demonstrated with a 3.36-fold speedup in action-chunk generation, which is critical for real-time robotic applications. The inclusion of real-world experiments strengthens the claim of practical utility.
The paper provides a clear description of the PDTD mechanism and the training objective. However, as an arXiv preprint without a specified code release in the provided text, full reproducibility depends on the availability of the codebase, which is not explicitly linked in the abstract or header. The architectural details of the "action-expert" and the specific hyperparameters for the drifting objective would be necessary for exact replication.
The evaluation is limited to specific benchmarks (LIBERO, RoboTwin) and a small set of real-world tasks. It is unclear how the method generalizes to more complex, long-horizon tasks or different robot embodiments beyond the tested arms. The "native one-step" approach may sacrifice some of the expressiveness or robustness of multi-step flow matching in highly stochastic or complex environments, though the paper claims no sacrifice in control performance. The lack of ablation studies on the specific impact of PDTD versus standard one-step distillation is a potential gap.
This work contributes to the trend of making VLA models more efficient and deployable in real-time settings. By eliminating iterative refinement, it lowers the computational barrier for high-frequency control, potentially enabling the use of larger VLA models on embedded hardware. The per-dimension decomposition strategy could inspire similar fine-grained modeling approaches in other continuous control domains. DriftingVLA introduces a native one-step VLA generation method using Per-Dimension Temporal Drifting to achieve significant latency reduction without sacrificing performance. The paper presents a solid engineering and algorithmic contribution to the VLA field, offering a practical solution to the inference latency problem inherent in flow-based models, with strong empirical results on both simulation and real-world robotic tasks.